in each category
(column or bar chart)
See Graphical presentation for descriptions of the main graphical techniques.
Mean – the arithmetic average, calculated by summing all the values and dividing by the number of values in the sum.
Median – the mid point of the distribution, where half the values are higher and half lower.
Mode – the most frequently occurring value.
Range – the difference between the highest and lowest value.
Inter-quartile range – the difference between the upper quartile (the value where 25 per cent of the observations are higher and 75 per cent lower) and the lower quartile (the value where 75 per cent of the observations are higher and 25 per cent lower). This is particularly useful where there are a small number of extreme observations much higher, or lower, than the majority.
Variance – a measure of spread, calculated as the mean of the squared differences of the observations from their mean.
Standard deviation – the square root of the variance.
Chi-squared test – used to compare the distributions of two or more sets of categorical or ordinal data.
t-tests – used to compare the means of two sets of data.
Wilcoxon U test – non-parametric equivalent of the t-test. Based on the rank order of the data, it may also be used to compare medians.
ANOVA – analysis of variance, to compare the means of more than two groups of data.
Compare two groups | Categorical | Chi-squared test |
Ordinal | Chi-squared test Wicoxon U test | |
Ratio / Interval | t-test for independent samples | |
Compare more than two groups | Categorical / Ordinal | Chi-squared test |
Ratio / Interval | ANOVA | |
Compare two variables over the same subjects | Categorical / Ordinal | Chi-squared test |
Ratio / Interval | t-test for dependent samples |
The correlation coefficient measures the degree of linear association between two variables, with a value in the range +1 to -1. Positive values indicate that the two variables increase and decrease together; negative values that one increases as the other decreases. A correlation coefficient of zero indicates no linear relationship between the two variables. The Spearman rank correlation is the non-parametric equivalent of the Pearson correlation.
Categorical | Chi-squared test |
Ordinal | Chi-squared test Spearman rank correlation (Tau) |
Ratio / Interval | Pearson correlation (Rho) |
Note that correlation analyses will only detect linear relationships between two variables. The figure below illustrates two small data sets where there are clearly relationships between the two variables. However, the correlation for the second data set, where the relationship is not linear, is 0.0. A simple correlation analysis of these data would suggest no relationship between the measures, when that is clearly not the case. This illustrates the importance of undertaking a series of basic descriptive analyses before embarking on analyses of the differences and relationships between variables.
The statistical significance of a test is a measure of probability - the probability that you would have obtained that particular result of the test on that sample if the null hypothesis (that there is no effect due to the parameters being tested) you are testing was true. The example below tests whether scores in an exam change after candidates have received training. The hypothesis suggests that they should, so the null hyopothesis is that they won't .
In general, any level of probability above 5 per cent (p>0.05) is not considered to be statistically significant, and for large surveys 1 per cent (p>0.01) is often taken as a more appropriate level.
Note that statistical significance does not mean that the results you have obtained actually have value in the context of your research. If you have a large enough sample, a very small difference between groups can be identified as statistically significant, but such a small difference may be irrelevant in practice. On the other hand, an apparently large difference may not be statistically significant in a small sample, due to the variation within the groups being compared.
Some test statistics (e.g. chi-squared) require the number of degrees of freedom to be known, in order to test for statistical significance against the correct probability table. In brief, the degrees of freedom is the number of values which can be assigned arbitrarily within the sample.
In a sample of size n divided into k classes, there are k-1 degrees of freedom (the first k-1 groups could be of any size up to n, while the last is fixed by the total of the first k-1 and the value of n. In numerical terms, if a sample of 500 individuals is taken from the UK, and it is observed that 300 are from England, 100 from Scotland and 50 from Wales, then there must be 50 from Northern Ireland. Given the numbers from the first three groups, there is no flexibility in the size of the final group. Dividing the sample into four groups gives three degrees of freedom.
In a two-way contingency table with p rows and q columns, there are (p-1)*(q-1) degrees of freedom (given the values of the first rows and columns, the last row and column are constrained by the totals in the table)
If, as is generally the case, what matters is simply that the statistics for the populations are different, then it is appropriate to use the critical values for a two-tailed test.
If, however, you are only interested to find out if the statistic for population A has a larger value than that for population B, then a one-tailed test would be appropriate. The critical value for a one-tailed test is generally lower than for a two-tailed test, and should only be used if your research hypothesis is that population A has a greater value than population B, and it does not matter how different they are if population A has a value that is less than that for population B.
Null hypothesis – there is no difference in mean exam scores before and after training (i.e. training has no effect on the exam score) Alternative – there is a difference in the mean scores before and after training (i.e. training has an unspecified effect) Use a two-tail test
Null hypothesis – Training does not increase the mean score Alternative – Mean score increases after training Use a one-tail test , if there is an observed increase in mean score. (If there is an observed fall in scores, there is no need to test, as you cannot reject the null hypothesis.)
Null hypothesis – Training does not cause mean scores to fall Alternative – Mean score falls after training Use a one-tail test , if there is an observed fall in mean score. (If there is an observed increase in scores, there is no need to test, as you cannot reject the null hypothesis.)
Mean | 360.4 | 361.1 |
Variance | 46,547 | 46,830 |
Observations | 62 | 62 |
Degrees of freedom (df) | 61 | |
t Stat | 1.79 | |
| ||
t Critical one-tail | 1.67 | |
| ||
t Critical two-tail | 2.00 |
If the above test results were obtained, then under scenario 1, using a two-tail test, you might conclude that there was no statistically significant difference between the scores (p=0.08), and, as a consequence, that training had no effect. Similarly, under scenario 3, you would conclude that there is no evidence to suggest that training causes mean scores to fall, as they have in fact risen. However, under scenario 2, using a one-tail test, you would conclude that there was an increase in mean scores, statistically significant at the 5 per cent level (p=0.04).
Statistical packages will do what you tell them, on the whole. They do not know whether the data you have provided is of good quality, or (with a very few exceptions) whether it is of an appropriate type for the analysis you have undertaken.
Rubbish in = Rubbish out!
These tools and techniques have specialist applications, and will generally be designed into the research methodology at an early stage, before any data are collected. If you are considering using any of these, you may wish to consult a specialist text or an experienced statistician before you start.
In each case, we give some examples of Emerald articles which use the technique.
To reduce the number of variables for subsequent analysis by creating combinations of the original variables measured which account for as much of the original variance as possible, but allow for easier interpretation of the results. Commonly used to create a small set of dimension ratings from a large number of opinion statements individually rated on Likert scales. You must have more observations (subjects) than you have variables to be analysed.
A Likert scale variable: "I like to eat chocolate ice cream for breakfast"
Strongly agree | 1 | 2 | 3 | 4 | 5 | Strongly disagree |
A factor analysis of Page and Wong's servant leadership instrument Rob Dennis and Bruce E. Winston Leadership & Organization Development Journal , vol. 24 no. 8
Understanding factors for benchmarking adoption: New evidence from Malaysia Yean Pin Lee, Suhaiza Zailani and Keng Lin Soh Benchmarking: An International Journal , vol. 13 no. 5
To classify subjects into groups with similar characteristics, according to the values of the variables measured. You must have more observations than you have variables included in the analysis.
Organic product avoidance: Reasons for rejection and potential buyers' identification in a countrywide survey C. Fotopoulos and A. Krystallis British Food Journal , vol. 104 no. 3/4/5
Detection of financial distress via multivariate statistical analysis S. Gamesalingam and Kuldeep Kumar Managerial Finance , vol. 27 no. 4
To identify those variables which best discriminate between known groups of subjects. The results may be used to allocate new subjects to the known groups based on their values of the discriminating variables
Methodology
Discriminant analysis was used to determine whether statistically significant differences exist between the average score profile on a set of variables for two a priori defined groups and so enabled them to be classified. Besides, it could help to determine which of the independent variables account the most for the differences in the average score profiles of the two groups. In this study, discriminant analysis was the main instrument to classify the benchmarking adopter and non-adopter. It was also utilised to determine which of the independent variables would contribute to benchmarking adoption.
To model how one, dependant, variable behaves depending on the values of a set of other, independent, variables. The dependant variable must be interval or ratio in type; the independent variables may be of any type, but special methods must be used when including categorical or ordinal independent variables in the analysis.
Developments in milk marketing in England and Wales during the 1990s Jeremy Franks British Food Journal , vol. 103 no. 9
Training under fire: The relationship between obstacles facing training and SMEs' development in Palestine Mohammed Al Madhoun Journal of European Industrial Training , vol. 30 no. 2
To investigate the patterns and trends in a variable measured regularly over a period of time. May also be used to identify and adjust for seasonal variation, for example in financial statistics.
An analysis of the trends and cyclical behaviours of house prices in the Asian markets Ming-Chi Chen, Yuichiro Kawaguchi and Kanak Patel Journal of Property Investment & Finance , vol. 22 no. 1
Presenting data in graphical form can increase the accessibility of your results to a non-technical audience, and highlight effects and results which would otherwise require lengthy explanation, or complex tables. It is therefore important that appropriate graphical techniques are used. This section gives examples of some of the most commonly used graphical presentations, and indicates when they may be used. All, except the histogram, have been produced using Microsoft Excel®.
There are four main variations, and whether you display the data in horizontal bars or vertical columns is largely a matter of personal preference.
To illustrate a frequency distribution in categorical or ordinal data, or grouped ratio/interval data. Usually displayed as a column graph.
To compare categorical, ordinal or grouped ratio/interval data across categories. The data used in fig 4 are the same as those in Figs 5 and 6.
To illustrate the actual contribution to the total for categorical, ordinal or grouped ratio/interval data by categories. The data used in Fig 5 are the same as those in Figs 4 and 6.
To compare the percentage contribution to the total for categorical, ordinal or grouped ratio/interval data across categories. The data used in fig 6 are the same as those in Figs 4 and 5.
To show trends in ordinal or ratio/interval data. Points on a graph should only be joined with a line if the data on the x-axis are at least ordinal. One particular application is to plot a frequency distribution for interval/ratio data (fig 8).
To show the percentage contribution to the whole of categorical, ordinal or grouped ratio/interval data.
To illustrate the relationship between two variables, of any type (although most useful where both variables are ratio/interval in type). Also useful in the identification of any unusual observations in the data.
A specialist graph illustrating the central tendency and spread of a large data set, including any outliers.
Connecting Mathematics Brief explanations of mathematical terms and ideas
Statistics Glossary compiled by Valerie J. Easton and John H. McColl of Glasgow University
Statsoft electronic textbook
100 Statistical Tests by Gopal K. Kanji (Sage, 1993, ISBN 141292376X)
Oxford Dictionary of Statistics by Graham Upton and Ian Cook (Oxford University Press, 2006, ISBN 0198614314)
Discover the world's research
Introduction.
Statistics is a field of science concerned with gathering, organising, analysing, and extrapolating data from samples to the entire population. This necessitates a well-designed study, a well-chosen study sample, and a proper statistical test selection. A good understanding of statistics is required to design epidemiological research or a clinical trial. Improper statistical approaches might lead to erroneous findings and unethical behaviour.
A variable is a trait that differs from one person to the next within a population. Quantitative variables are measured by a scale and provide quantitative information, such as height and weight. Qualitative factors, such as sex and eye colour, provide qualitative information (Figure 1).
Figure 1. Classification of variables [1]
Quantitative variables
Discrete and continuous measures are used to split quantitative or numerical data. Continuous data can take on any value, whereas discrete numerical data is stored as a whole number such as 0, 1, 2, 3,… (integer). Discrete data is made up of countable observations, while continuous data is made up of measurable observations. Discrete data examples include the number of respiratory arrest episodes or re-intubation in an intensive care unit. Continuous data includes serial serum glucose levels, partial pressure of oxygen in arterial blood, and oesophageal temperature. A hierarchical scale with increasing precision can be used based on category, ordinal, interval and ratio scales (Figure 1).
Descriptive statistics try to explain how variables in a sample or population are related. The mean, median, and mode forms, descriptive statistics give an overview of data. Inferential statistics use a random sample of data from that group to characterise and infer about a community as a whole. It’s useful when it’s not possible to investigate every single person in a group.
Descriptive statistics
The central tendency describes how observations cluster about a centre point, whereas the degree of dispersion describes the spread towards the extremes.
Inferential statistics
In inferential statistics, data from a sample is analysed to conclude the entire population. The goal is to prove or disprove the theories. A hypothesis is a suggested explanation for a phenomenon (plural hypotheses). Hypothesis testing is essential to process for making logical choices regarding observed effects’ veracity.
SOFTWARES FOR STATISTICS, SAMPLE SIZE CALCULATION AND POWER ANALYSIS
There are several statistical software packages accessible today. The most commonly used software systems are Statistical Package for the Social Sciences (SPSS – manufactured by IBM corporation), Statistical Analysis System (SAS – developed by SAS Institute North Carolina, Minitab (developed by Minitab Inc), United States of America), R (designed by Ross Ihaka and Robert Gentleman from the R core team), Stata (developed by StataCorp), and MS Excel. There are several websites linked to statistical power studies. Here are a few examples:
A researcher must be familiar with the most important statistical approaches for doing research. This will aid in the implementation of a well-designed study that yields accurate and valid data. Incorrect statistical approaches can result in erroneous findings, mistakes, and reduced paper’s importance. Poor statistics can lead to poor research, which can lead to immoral behaviour. As a result, proper statistical understanding and the right application of statistical tests are essential. A thorough understanding of fundamental statistical methods will go a long way toward enhancing study designs and creating high-quality medical research that may be used to develop evidence-based guidelines.
[1] Ali, Zulfiqar, and S Bala Bhaskar. “Basic statistical tools in research and data analysis.” Indian journal of anaesthesia vol. 60,9 (2016): 662-669. doi:10.4103/0019-5049.190623
[2] Ali, Zulfiqar, and S Bala Bhaskar. “Basic statistical tools in research and data analysis.” Indian journal of anaesthesia vol. 60,9 (2016): 662-669. doi:10.4103/0019-5049.190623
Statswork is a pioneer statistical consulting company providing full assistance to researchers and scholars. Statswork offers expert consulting assistance and enhancing researchers by our distinct statistical process and communication throughout the research process with us.
– Research Planning – Tool Development – Data Mining – Data Collection – Statistics Coursework – Research Methodology – Meta Analysis – Data Analysis
#10, Kutty Street, Nungambakkam, Chennai, Tamil Nadu – 600034, India No : +91 4433182000, UK No : +44-1223926607 , US No : +1-9725029262 Email: [email protected]
Website: www.statswork.com
© 2024 Statswork. All Rights Reserved
Warning: The NCBI web site requires JavaScript to function. more...
An official website of the United States government
The .gov means it's official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you're on a federal government site.
The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.
NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2024 Jan-.
Exploratory data analysis: frequencies, descriptive statistics, histograms, and boxplots.
Jacob Shreffler ; Martin R. Huecker .
Last Update: November 3, 2023 .
Researchers must utilize exploratory data techniques to present findings to a target audience and create appropriate graphs and figures. Researchers can determine if outliers exist, data are missing, and statistical assumptions will be upheld by understanding data. Additionally, it is essential to comprehend these data when describing them in conclusions of a paper, in a meeting with colleagues invested in the findings, or while reading others’ work.
This comprehension begins with exploring these data through the outputs discussed in this article. Individuals who do not conduct research must still comprehend new studies, and knowledge of fundamentals in analyzing data and interpretation of histograms and boxplots facilitates the ability to appraise recent publications accurately. Without this familiarity, decisions could be implemented based on inaccurate delivery or interpretation of medical studies.
Frequencies and Descriptive Statistics
Effective presentation of study results, in presentation or manuscript form, typically starts with frequencies and descriptive statistics (ie, mean, medians, standard deviations). One can get a better sense of the variables by examining these data to determine whether a balanced and sufficient research design exists. Frequencies also inform on missing data and give a sense of outliers (will be discussed below).
Luckily, software programs are available to conduct exploratory data analysis. For this chapter, we will be examining the following research question.
RQ: Are there differences in drug life (length of effect) for Drug 23 based on the administration site?
A more precise hypothesis could be: Is drug 23 longer-lasting when administered via site A compared to site B?
To address this research question, exploratory data analysis is conducted. First, it is essential to start with the frequencies of the variables. To keep things simple, only variables of minutes (drug life effect) and administration site (A vs B) are included. See Image. Figure 1 for outputs for frequencies.
Figure 1 shows that the administration site appears to be a balanced design with 50 individuals in each group. The excerpt for minutes frequencies is the bottom portion of Figure 1 and shows how many cases fell into each time frame with the cumulative percent on the right-hand side. In examining Figure 1, one suspiciously low measurement (135) was observed, considering time variables. If a data point seems inaccurate, a researcher should find this case and confirm if this was an entry error. For the sake of this review, the authors state that this was an entry error and should have been entered 535 and not 135. Had the analysis occurred without checking this, the data analysis, results, and conclusions would have been invalid. When finding any entry errors and determining how groups are balanced, potential missing data is explored. If not responsibly evaluated, missing values can nullify results.
After replacing the incorrect 135 with 535, descriptive statistics, including the mean, median, mode, minimum/maximum scores, and standard deviation were examined. Output for the research example for the variable of minutes can be seen in Figure 2. Observe each variable to ensure that the mean seems reasonable and that the minimum and maximum are within an appropriate range based on medical competence or an available codebook. One assumption common in statistical analyses is a normal distribution. Image . Figure 2 shows that the mode differs from the mean and the median. We have visualization tools such as histograms to examine these scores for normality and outliers before making decisions.
Histograms are useful in assessing normality, as many statistical tests (eg, ANOVA and regression) assume the data have a normal distribution. When data deviate from a normal distribution, it is quantified using skewness and kurtosis. [1] Skewness occurs when one tail of the curve is longer. If the tail is lengthier on the left side of the curve (more cases on the higher values), this would be negatively skewed, whereas if the tail is longer on the right side, it would be positively skewed. Kurtosis is another facet of normality. Positive kurtosis occurs when the center has many values falling in the middle, whereas negative kurtosis occurs when there are very heavy tails. [2]
Additionally, histograms reveal outliers: data points either entered incorrectly or truly very different from the rest of the sample. When there are outliers, one must determine accuracy based on random chance or the error in the experiment and provide strong justification if the decision is to exclude them. [3] Outliers require attention to ensure the data analysis accurately reflects the majority of the data and is not influenced by extreme values; cleaning these outliers can result in better quality decision-making in clinical practice. [4] A common approach to determining if a variable is approximately normally distributed is converting values to z scores and determining if any scores are less than -3 or greater than 3. For a normal distribution, about 99% of scores should lie within three standard deviations of the mean. [5] Importantly, one should not automatically throw out any values outside of this range but consider it in corroboration with the other factors aforementioned. Outliers are relatively common, so when these are prevalent, one must assess the risks and benefits of exclusion. [6]
Image . Figure 3 provides examples of histograms. In Figure 3A, 2 possible outliers causing kurtosis are observed. If values within 3 standard deviations are used, the result in Figure 3B are observed. This histogram appears much closer to an approximately normal distribution with the kurtosis being treated. Remember, all evidence should be considered before eliminating outliers. When reporting outliers in scientific paper outputs, account for the number of outliers excluded and justify why they were excluded.
Boxplots can examine for outliers, assess the range of data, and show differences among groups. Boxplots provide a visual representation of ranges and medians, illustrating differences amongst groups, and are useful in various outlets, including evidence-based medicine. [7] Boxplots provide a picture of data distribution when there are numerous values, and all values cannot be displayed (ie, a scatterplot). [8] Figure 4 illustrates the differences between drug site administration and the length of drug life from the above example.
Image . Figure 4 shows differences with potential clinical impact. Had any outliers existed (data from the histogram were cleaned), they would appear outside the line endpoint. The red boxes represent the middle 50% of scores. The lines within each red box represent the median number of minutes within each administration site. The horizontal lines at the top and bottom of each line connected to the red box represent the 25th and 75th percentiles. In examining the difference boxplots, an overlap in minutes between 2 administration sites were observed: the approximate top 25 percent from site B had the same time noted as the bottom 25 percent at site A. Site B had a median minute amount under 525, whereas administration site A had a length greater than 550. If there were no differences in adverse reactions at site A, analysis of this figure provides evidence that healthcare providers should administer the drug via site A. Researchers could follow by testing a third administration site, site C. Image . Figure 5 shows what would happen if site C led to a longer drug life compared to site A.
Figure 5 displays the same site A data as Figure 4, but something looks different. The significant variance at site C makes site A’s variance appear smaller. In order words, patients who were administered the drug via site C had a larger range of scores. Thus, some patients experience a longer half-life when the drug is administered via site C than the median of site A; however, the broad range (lack of accuracy) and lower median should be the focus. The precision of minutes is much more compacted in site A. Therefore, the median is higher, and the range is more precise. One may conclude that this makes site A a more desirable site.
Ultimately, by understanding basic exploratory data methods, medical researchers and consumers of research can make quality and data-informed decisions. These data-informed decisions will result in the ability to appraise the clinical significance of research outputs. By overlooking these fundamentals in statistics, critical errors in judgment can occur.
All interprofessional healthcare team members need to be at least familiar with, if not well-versed in, these statistical analyses so they can read and interpret study data and apply the data implications in their everyday practice. This approach allows all practitioners to remain abreast of the latest developments and provides valuable data for evidence-based medicine, ultimately leading to improved patient outcomes.
Exploratory Data Analysis Figure 1 Contributed by Martin Huecker, MD and Jacob Shreffler, PhD
Exploratory Data Analysis Figure 2 Contributed by Martin Huecker, MD and Jacob Shreffler, PhD
Exploratory Data Analysis Figure 3 Contributed by Martin Huecker, MD and Jacob Shreffler, PhD
Exploratory Data Analysis Figure 4 Contributed by Martin Huecker, MD and Jacob Shreffler, PhD
Exploratory Data Analysis Figure 5 Contributed by Martin Huecker, MD and Jacob Shreffler, PhD
Disclosure: Jacob Shreffler declares no relevant financial relationships with ineligible companies.
Disclosure: Martin Huecker declares no relevant financial relationships with ineligible companies.
This book is distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) ( http://creativecommons.org/licenses/by-nc-nd/4.0/ ), which permits others to distribute the work, provided that the article is not altered or used commercially. You are not required to obtain permission to distribute this article, provided that you credit the author and journal.
Bulk download.
Your browsing activity is empty.
Activity recording is turned off.
Turn recording back on
Connect with NLM
National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894
Web Policies FOIA HHS Vulnerability Disclosure
Help Accessibility Careers
You have full access to this open access article
This paper highlights the scarring effects of early life exposure to civil war, by examining the impact of exposure to conflict in childhood on the incidence of domestic violence in adulthood among married women. To estimate these effects, we use a difference-in-differences model which exploits variation in exposure to Nigeria’s 30-month-long civil war by year of birth and ethnicity. Our results, based on the 2008 Nigerian Demographic Health Survey, show that women exposed to the war during childhood are more likely to be victims of domestic violence in adulthood compared to those not exposed to the war, with larger effects observed for those exposed at younger ages. Additionally, we explore the mechanisms through which exposure to civil war might affect domestic violence and find some support for both the normalisation of violence and weakened bargaining power hypotheses. Understanding the root causes of domestic violence is important given the high prevalence in developing countries and the deleterious consequences for women and their children.
Ce document met en évidence les effets cicatrisants d'une exposition précoce à la guerre civile, en examinant l'impact de l'exposition au conflit pendant l'enfance sur l'incidence de la violence domestique à l'âge adulte chez les femmes mariées. Pour estimer ces effets, nous utilisons un modèle de différences en différences qui exploite la variation de l'exposition à la guerre civile nigériane de 30 mois en fonction de l'année de naissance et de l'ethnicité. Nos résultats, basés sur l'Enquête démographique de santé nigériane de 2008, montrent que les femmes exposées à la guerre pendant l'enfance sont plus susceptibles d'être victimes de violence domestique à l'âge adulte par rapport à celles qui n'ont pas été exposées à la guerre, avec des effets plus importants observés pour celles exposées à des âges plus jeunes. De plus, nous explorons les mécanismes par lesquels l'exposition à la guerre civile pourrait affecter la violence domestique et trouvons un certain soutien pour les hypothèses de normalisation de la violence et d'affaiblissement du pouvoir de négociation. Comprendre les causes profondes de la violence domestique est important étant donné la prévalence élevée dans les pays en développement et les conséquences délétères pour les femmes et leurs enfants.
Este documento destaca los efectos perjudiciales de la exposición en los primeros años de vida a la guerra civil, examinando el impacto de la exposición al conflicto en la infancia sobre la incidencia de la violencia doméstica en la adultez entre mujeres casadas. Para estimar estos efectos, utilizamos un modelo de diferencias en diferencias que explota la variación en la exposición a la guerra civil de Nigeria de 30 meses de duración por año de nacimiento y etnia. Nuestros resultados, basados en la Encuesta de Salud Demográfica de Nigeria 2008, muestran que las mujeres expuestas a la guerra durante la infancia tienen más probabilidades de ser víctimas de violencia doméstica en la adultez en comparación con aquellas que no estuvieron expuestas a la guerra, con efectos mayores observados para aquellas expuestas a edades más tempranas. Además, exploramos los mecanismos a través de los cuales la exposición a la guerra civil podría afectar la violencia doméstica y encontramos cierto apoyo tanto para las hipótesis de normalización de la violencia como para el debilitamiento del poder de negociación. Comprender las causas fundamentales de la violencia doméstica es importante dado su alta prevalencia en los países en desarrollo y las consecuencias perjudiciales para las mujeres y sus hijos.
Avoid common mistakes on your manuscript.
Since World War II, almost one-third of all countries have experienced civil war, and the incidence of armed conflict has been on the rise (Gleditsch et al. 2002 ). In Sub-Saharan Africa specifically, nearly three-fourths of countries in the region have experienced civil war (Gleditsch et al. 2002 ). These conflicts have often led to considerable loss of lives, deterioration of physical and human capital, erosion of institutional capacity, and reduced economic growth (Akbulut-Yuksel and Yuksel 2017 ). It has been estimated, for instance, that between 2012 and 2017, the global economic costs of conflict increased from $12.62 trillion to $14.76 trillion, with many of the conflict-torn countries trapped in a perpetual cycle of violence (World Development Report 2011 ; World Humanitarian Data and Trends Report 2017 ; Institute for Economics and Peace 2018 ).
While the macroeconomic costs of war have long been studied in economics, literature on the microeconomic impacts of civil war, particularly in developing countries, has grown in the last 20 years especially, perhaps as more data have become available (Verwimp et al 2019 ). Studies have shown that exposure to conflict is negatively associated with educational attainment (Singh and Shemyakina 2016 ; Chamarbagwala and Moran 2011 ; Shemyakina 2011 ; Swee 2015 ), health outcomes (Akresh et al. 2012a , 2012b ; Grimard and Laszlo 2014 ; Weldeegzie; 2017 ), social trust (Kijewski and Freitag 2018 ), and labour market outcomes (Galdo 2013 ; Islam et al. 2016 ).
In this paper, we add to this literature by exploring how exposure to conflict in childhood affects experiences of domestic violence among women in adulthood, using the case of the Nigerian civil war. Recent work suggests that exposure to war increases women’s likelihood of experiencing intimate partner violence across a range of contexts. La Mattina ( 2017 ) finds that exposure to the genocide in Rwanda increased the incidence of domestic violence among women who married after 1994 compared to those who married before the genocide occurred, with a larger effect for women in areas with high genocide intensity. Kelly et al ( 2018 ) match district-level information on conflict-related fatalities during the civil war in Liberia from 1999 to 2003 to data on post-conflict intimate partner violence from the 2007 Demographic Health Survey (DHS). They find a strong effect of fatalities on the incidence of intimate partner violence, with 4–5 years of cumulative exposure having the strongest effect. In a similar vein, Østby et al ( 2019 ) analyse the experiences of women in Peru during and after the civil war from 1980 to 2000 and find that those living in areas with higher exposure to conflict-related violence are at increased risk of violence in the home. Svallfors ( 2023 ) analyses DHS data from 2005 to 2015 for Columbia and shows that local-level exposure to armed conflict events in the previous year especially, increased women’s likelihood of experiencing intimate partner violence.
In all these studies, the focus has been on the association between conflict exposure and domestic violence in adulthood, or on temporally proximate relationships. In our reading of the literature, we could find only one very recent published paper by Torrisi ( 2023 ) which tries to uncover whether the timing of exposure matters, and particularly whether exposure to armed conflict during childhood has long-lasting consequences for domestic violence in adulthood. Torrisi ( 2023 ) combines DHS data with geo-referenced information on the armed conflicts that occurred in four ex-Soviet countries (Armenia, Azerbaijan, Moldova, and Tajikistan) soon after the break-up of the USSR. She finds that women who were exposed to conflict by age 19 were more likely to experience domestic violence than those never exposed or not exposed by age 19, and that this effect is driven largely by exposure in the sensitive childhood period from 0 to 10 years of age (with no significant effect for those exposed at ages 11 to 15 or 16 to 19).
We also found two working papers that explore the relationship between childhood exposure and domestic violence in adulthood (Gutierrez and Gallegos 2016 ; La Mattina and Shemyakina 2017 ). Gutierrez and Gallegos ( 2016 ) use DHS data from Peru coupled with information on geographical variation in exposure to violent conflict to show that both women who were exposed at ages 0 to 8 and 9 to 16 experienced a higher incidence of domestic violence in adulthood compared to those not exposed. La Mattina and Shemyakina ( 2017 ) use the DHS data on selected Sub-Saharan African countries and exploit both temporal and geographical variation in conflict intensity between 1946 and 2006 across sub-national regions. Their results suggest that women who live in a region where there was an armed conflict when they were 6 to 10 years old are more likely to experience domestic violence than individuals not exposed to conflict by age 20, but they do not observe similar effects for individuals who were exposed to conflict at ages 0 to 5 or 11 to 20.
There is a common methodological thread that runs throughout all these studies: they use geo-referenced data on conflict-related violence combined with post-conflict data on domestic violence from the DHS surveys. In addition to imperfect matching at the sub-national or district level due to differences in levels of geographical disaggregation or demarcation between the two sources of data, a key concern with this approach is endogenous migration. The DHS only has information on the individual’s current place of residence and not on their residence in childhood or at the time of conflict. There is therefore no guarantee that the women who are currently living in a previously conflict-exposed area were also living there during childhood when the conflict took place. Indeed, endogenous migration is likely to be more of a concern during times of conflict, and the direction of the effect is difficult to predict. It is possible that the most vulnerable women (and men) may be displaced or forced to flee with their families during times of conflict, but it is also possible that the least vulnerable, those with better economic resources and social networks, are the ones who can more easily relocate to places of safety. To try and address this problem, many of the studies listed above restrict their samples to those who had never moved since birth or who had not moved in the previous five years, depending on the data available in the DHS. In doing so, however, they tend to lose 50 percent or more of their initial sample (Gutierrez and Gallegos 2016 ; La Mattina and Shemyakina 2017 ; Torrisi 2023 ), likely leading to biassed results.
Our paper makes a useful methodological contribution to this growing literature on the long-term effects of war exposure by using what we consider to be a more robust method of identifying exposure than the commonly used geographical approach. We use ethnicity and birth cohort to identify exposure to conflict in childhood during the Nigerian civil war (following the approach adopted in Akresh et al 2012a , 2023 ). We are able to adopt this approach because of the very specific nature of the Nigerian civil war, which occurred from 6 July 1967 to 15 January 1970, and which was restricted to the south-eastern region of Nigeria inhabited by the Igbos and other minority ethnic groups (which we will describe in more detail below). This strategy mitigates the problem of selective migration associated with the use of geography-based variables to identify exposure, a problem which is likely to be more pronounced during times of conflict.
In addition, we examine exposure in early childhood using more granular age ranges than have currently been explored, namely those exposed in utero, between the ages of 0 to 4, 5 to 8, and 9 to 12. In doing so, we add to the growing body of literature in economics which recognises that there are long-run implications of early life shocks and that adverse circumstances during the sensitive early period of childhood can impact a range of later life outcomes (Case et al. 2005 ; Cunha and Heckman 2007 ; Almond and Currie 2011 ; Currie 2020 ). This includes increasing evidence that in utero exposure to shocks such as war, disease, and famine have long-term negative consequences on physical and mental health, educational attainment, earnings, and other socio-economic outcomes (Almond 2006 ; Camacho 2009 ; Almond and Currie 2011 ; Comfort 2016 ; Almond et al. 2018 ).
Finally, we try to unpack the mechanisms through which early life exposure to conflict affects experiences of domestic violence in adulthood, using the rich data available in the Nigerian Demographic Health Survey. We explore two possible channels. The first, the normalisation of violence hypothesis, relies on the well-known finding that children who witness violence at home are more likely to become a victim or perpetrator of domestic violence themselves in adulthood (Schwab-Stone et al. 1995 ; Gage 2005 ; Yount and Li 2009 ; Cesur and Sabia 2016 ; Jin et al. 2017 ). If war results in more intimate partner violence among married couples, as the evidence presented earlier suggests, we would expect children growing up during war to witness more violence among their parents than observably similar children. Even if children do not witness violence within their own homes, one might expect that children exposed to community-level violence through war during their formative years might also be more likely to view violence as a justifiable response to certain problems (Barnett et al. 2005 ; Fowler et al. 2009 ; Gutierrez and Galegos 2016 ). To examine whether exposure to violence in childhood might have affected the formation of beliefs during the critical early years, we use data in the DHS on whether war-exposed women witnessed domestic violence in their homes as children and on women’s and men’s attitudes towards wife-beating in adulthood (Huber 2023 ).
The second hypothesis we explore is reduced bargaining power in the household, which would affect women’s options outside of the marriage and in turn increase their likelihood of being victims of domestic violence (Bhattacharyya et al. 2011 ; Heath 2014 ; La Mattina 2017 ). There are a number of reasons why women exposed to war may have fewer outside options. For instance, a number of studies in a range of countries have found evidence that civil conflict results in poorer educational outcomes (Akresh and Walque 2008 ; Leon 2012 ; Shemyakina 2011 ; Chamarbagwala and Moran 2011 ; and Dabalen and Paul 2014 ), and there is some evidence that exposure to conflict negatively affects girls more than boys (Singh and Shemyakina 2016 ). Women with lower education have fewer out-of-marriage options given their weaker labour market outcomes and increased financial dependence on their husbands, raising the likelihood of domestic violence (Lundberg and Pollak 1996 ; Farmer and Tiefenthaler 1997 ; Aizer 2010 ; Bhattacharyya et al. 2011 ; Eswaran and Malhotra 2011 ; Galdo 2013 ; Heath 2014 ). Moreover, war exposure can affect marriage, reproductive and health outcomes, which would have consequences for women’s intra-household bargaining power (Verwimp and van Bavel 2005 ; Aizer 2011 ; Akresh 2012a ; Islam et al 2016 ; Cetorelli and Khawaja 2017 ; La Mattina 2017 ). To measure women’s bargaining power in adulthood, we use the information in the DHS on women’s decision-making power in the household across a number of domains (Ajefu and Casale 2021 ).
Our main findings are as follows. We find that women exposed to the Nigerian civil war during childhood are more likely to be victims of domestic violence in adulthood compared to women not exposed to the civil war. Specifically, we find that exposure to the civil war is associated with an increase in the likelihood of being a victim of domestic violence of 1.2 percentage points compared to non-exposed cohorts (or 6% given the sample mean incidence of 19.7%). These effects appear to be more pronounced the earlier on one is exposed in childhood, with particularly large effects for those exposed in utero. While it is far more difficult to identify the channels through which exposure to the civil war affects domestic violence (particularly across the cohorts), in our exploratory work, we find some evidence to support both the normalisation of violence and bargaining power hypotheses.
The rest of the paper is structured as follows. Section 2 provides background information on the Nigerian civil war. Section 3 discusses the data and the empirical identification strategy, and presents some descriptive statistics. Section 4 presents the estimation results, and Sect. 5 concludes.
Under British colonial rule, Nigeria comprised three regions, namely the northern, western, and eastern regions. Footnote 1 Each of these regions had a predominant ethnic group, with the Hausa in the North, the Yoruba in the West, and the Igbo in the East. Like many countries in Africa, political and social conflict in Nigeria bore both ethnic and regional dimensions (Simpson 2014 ). In less than seven years after becoming an independent nation (on 1 October 1960), some of these long-standing tensions between the different groups intensified and the country was plunged into a civil war, also known as the Biafran War.
While the underlying geo-political causes of the war are too complex to explain here, some of the immediate causes of the Nigerian Civil War were the military coup on 15 January 1966, organised by primarily Igbo army officers, the counter-coup of 28 July 1966, and the subsequent persecution and killing of the Igbos in the Northern part of the country (Kirk-Greene 1971 ; Nafziger 1972 ). In response to this, there was a massive return migration of Igbos seeking refuge (estimated to involve around 1.5 million people) to their homeland in the south-eastern region (Aall 1970 ; Akresh et al 2012a ). On 30 May 1967, the south-eastern region declared itself the Republic of Biafra and this led to a full-blown civil war that began on 6 July 1967 (see Fig. 1 ).
Map of Nigeria indicating the south-east states. The civil war was restricted to the south-east region that declared itself the Biafra republic
Nigeria’s Federal Military Government fiercely resisted the breakaway republic for two and a half years, using both their military might and their ability to impose a blockade of the landlocked territory (preventing the inflow of food, medicine, and other essential supplies). It has been estimated that between 1 and 3 million people died from the violence and mass starvation that ensued, in what was considered one of the bloodiest wars in sub-Saharan Africa (Akresh et al. 2012a ; Simpson 2014 ). The war ended on 15 January 1970 after the Republic of Biafra surrendered to the Nigerian troops.
Two key features of this devastating conflict are salient to our empirical strategy. First, because of the military blockade (which prevented movement of both people and supplies), the war was fought in the south-eastern region with direct civilian exposure largely restricted to this area (Akresh et al. 2012a ). Second, at the time of the war, most Igbos were living in their native states in the south-east, and many of those living outside the area returned there before the war to seek refuge in the mass migration that occurred just before secession was declared (Aall 1970 ). We can therefore use ethnicity and birth cohort to identify exposure to the civil war. This identification strategy is similar to that used by Akresh et al ( 2012a ) in their study on the impact of exposure to the Nigerian civil war on women’s stature in adulthood. This strategy is preferred to using current geographical demarcation, as is the case in other studies exploring the relationship between war exposure and domestic violence, as it circumvents the problem of selective migration (ethnicity is invariant to migration).
To investigate the impact of the Nigerian civil war on women’s experience of domestic violence in adulthood, we use the 2008 Nigerian Demographic Health Survey (DHS). The DHS is a large nationally representative cross-sectional survey conducted in a number of developing countries. It provides information on women between the ages of 15 and 49 years on a large number of demographic and socio-economic factors. The 2008 Nigerian DHS covered 34,070 households and 33,385 women. Footnote 2 We use the 2008 survey in this study for two main reasons: it is the first wave of the Nigerian DHS to collect information on the incidence of domestic violence among women; and given the timing of the war, this particular survey covers the largest sample of war-exposed women, allowing us to explore the effects of exposure in utero through to exposure at 12 years of age. Footnote 3
The information on domestic violence was collected through a specially designed questionnaire that was administered to one randomly selected woman in each household. Footnote 4 Women who were (or had been) married or cohabiting were asked in private about incidents of domestic violence as follows: “(Does/did) your (last) husband ever do any of the following things to you: (a) slap you? (b) twist your arm or pull your hair? (c) push you, shake you, or throw something at you? (d) punch you with his fist or with something that could hurt you? (e) kick you, drag you or beat you up? (f) try to choke you or burn you on purpose? (g) threaten or attack you with a knife, gun, or any other weapon? (h) physically force you to have sexual intercourse with him even when you did not want to? (i) force you to perform any sexual acts you did not want to?” We measure domestic violence using a binary variable that takes the value of 1 if a woman suffered any of the above-mentioned aggressive behaviours from her husband or partner and 0 otherwise.
To estimate the causal impact of exposure to the civil war in childhood on experiences of domestic violence in adulthood, we adopt a difference-in-differences strategy. As described above, our identification strategy exploits variation in exposure to the civil war by birth cohort and ethnicity. This estimation strategy minimises the problem of selective migration associated with the use of geographical variation in conflict exposure and helps to circumvent one of the limitations of the Nigerian DHS, namely, that it only has information on the current residence of respondents but no information on their place of birth or their place of residence during the war.
We define the treatment or war-exposed group as those Igbo and other minority ethnic groups (who would have been in the south-eastern region when the war was fought) born between 1958 and October 1970. These women were between 0 and 12 years old (including in utero) when the war took place between July 1967 and January 1970, and are aged 38 to 49 years in 2008 when we observe their experiences of domestic violence.
We present two distinct control groups: i) one across time, i.e. women from the war-exposed ethnicities but born in the six-year period following the war, namely from November 1970 to December 1976 (and aged 32 to 38 years in 2008), Footnote 5 and ii) one across ethnicity, i.e. the same birth cohorts (1958–1976) but from the non-war-exposed ethnicities (predominant in the other regions of Nigeria). Table 1 summarises birth cohorts for the war-exposed and non-exposed groups, respectively.
We estimate Eq. ( 1 ) below:
where \({\text{Y}}_{\text{ijt}}\) is equal to one (zero otherwise) if individual i belonging to ethnicity j and born in year t was a victim of domestic violence in adulthood. \(wa{r}_{ethnicity}\) denotes Igbo or other minority ethnic groups in the south-east region and \({Cohort}_{it}\) includes four cohorts, namely those exposed to war in utero (born between February and October 1970), those exposed between 0 and 4 years (born 1966–1970), those exposed between 5 and 8 years (born 1962–1965), and those exposed between 9 and 12 years (born 1958–1961), where the omitted category is those born between November 1970 (i.e. nine months after the war) and December 1976. The interactions of war ethnicity with each of the four cohorts are the variables of interest and capture the effect of an individual’s exposure to the civil war on the incidence of domestic violence. \({X}_{ij}\) is a vector of individual and household characteristics, which includes age at first marriage, religion, education, urban residence, and household wealth; \({\delta }_{r}\) is a state fixed effect; and \({\varepsilon }_{ijt}\) is a random, idiosyncratic error term. We estimate the regressions using ordinary least squares (OLS) (although the results are robust to using probit regressions), and standard errors are clustered at the ethnicity level to account for serial correlation (Bertrand et al. 2004 ).
Table 2 reports the summary statistics for our sample of married/cohabiting women from whom domestic violence data were collected. The average age of women in this sample was 39 years, the average age at first marriage was 19 years, around 47% of women in the sample had completed at least primary education, and 32% were resident in urban areas. Among the women who were surveyed, 20% said they had experienced at least one type of domestic violence from their partner.
To explore the normalisation of violence and bargaining power hypotheses as potential mechanisms through which exposure to conflict affects the incidence of domestic violence, we also examine data on attitudes towards domestic violence, domestic violence among parents, and decision-making in the household. The summary statistics for these variables are also shown in Table 2 . On average, 34% of the women in the sample responded that domestic violence is justified if the woman goes out without informing the husband/partner, 32% felt it was justified if a woman neglects the children, 29% felt it was justified if a woman argues with her husband/partner, 26% felt it was justified if a woman refuses to have sex with her husband/partner, and 17% justified violence if a woman burns the food. Nearly 13% percent of women reported witnessing domestic violence among their own parents. In terms of household decision-making, 12% of women reported having the final say on own health care, 7% reported having the final say on large household purchases, 20% reported having the final say on household purchases for daily needs, and 14% reported having the final say on visits to family or relatives.
Table 3 shows that are large and significant differences in these variables by war exposure. Just under 18% of the non-exposed group reported being victims of domestic violence, compared to 27% of the war-exposed group. Moreover, 11% of the non-exposed group witnessed domestic violence among their parents, compared to 19% of the war-exposed group. There are also statistically significant differences in attitudes towards domestic violence, with war-exposed women more likely to report that wife-beating was justified in certain circumstances. For example, 15% of the non-exposed group justified wife-beating if a woman refuses to have sex with her partner compared to 30% of the war-exposed group. In terms of household decision-making, statistically significant differences are observed in three out of the four domains, with war-exposed women less likely to report having the final say on own health care, purchases for daily needs and visits to family and friends.
Figure 2 presents a box plot of our main variable of interest, the incidence of domestic violence, across the cohorts. Within each birth cohort, the incidence of domestic violence is clearly higher for the war-exposed ethnic groups compared to the non-exposed ethnic groups, and the difference between the two appears larger for those exposed at younger ages. However, these are unconditional estimates, and it remains to be seen whether these effects will hold in the multivariate difference-in-differences analysis, which we present in the next section.
Box plot showing the incidence of domestic violence across the cohorts for the exposed and non-exposed ethnicities
Table 4 presents the results from a series of equations which estimate the effect of exposure to the civil war in childhood (in utero to age 12) on the incidence of domestic violence in adulthood, without disaggregating by birth cohort. The coefficients on the interaction term suggest a positive and significant effect of war exposure in childhood on the incidence of domestic violence among women in adulthood. The size of the coefficient tends to fall as an increasing number of controls are added between columns 1 and 4. The regression in column 4 includes controls for individual and household characteristics and fixed effects for state, ethnicity, and cohort, and is our preferred specification. The coefficient from this regression suggests that exposure to the civil war increases the likelihood of being a victim of domestic violence by 1.2 percentage points (or 6% given the sample mean incidence of 19.7%). Footnote 6
In Table 5 , we disaggregate exposure to the civil war by birth cohort to test whether the effects of civil war exposure on domestic violence vary by the age at which the women were exposed to the war in childhood. The categories represent those exposed in utero (born between February 1970 and October 1970), those exposed between the ages of 0–4 (born 1966–1970), those exposed between the ages of 5–8 (born 1962–1965), and those exposed between the ages of 9–12 (born 1958–1961). From the estimates, we find that the effects are largest for those exposed at younger ages. Specifically, exposure to the civil war in utero increases the probability of experiencing domestic violence in adulthood by 7.4 percentage points, and exposure to the civil war between 0 and 4 years increases the probability of experiencing domestic violence by 1.7 percentage points (specification 4).
These results are consistent with the increasing evidence described earlier that there are long-run implications of early life shocks and that adverse circumstances during the sensitive early period of childhood impact later life outcomes (Case et al. 2005 ; Cunha and Heckman 2007 ; Currie 2020 ). This includes a growing body of literature showing that in utero exposure to shocks such as war, drought, and famine have long-term negative consequences.
This literature draws on the ‘fetal origins’ hypothesis, which proposes that conditions in utero, particularly nutrition, ‘program’ the foetus with particular metabolic features that can result in disease later on in life (Barker; 1990 , 1995 ). Studies have found evidence to link events or circumstances in utero to birth weight, adult height, disability, heart disease, and obesity, suggesting latent and long-lasting consequences on health outcomes (Ravelli et al 1976 ; Dunn 2007 ; Camacho 2009 ; Almond and Currie 2011 ; Comfort 2016 ). In addition, there is evidence to suggest negative effects on mental health and cognitive function as well as on education, employment, and adult earnings, implying potential neurological involvement (Hoek et al 1998 ; Almond 2006 ; Almond et al. 2018 ).
Almond et al ( 2018 ) summarise a number of ‘biological’ or direct mechanisms through which foetal-origin effects can be generated, including nutritional insults, infectious disease, maternal stress, and alcohol and tobacco use, all of which would likely be more prevalent during times of war. In addition to the direct biological mechanisms, there may be social and economic factors at play that reinforce the negative outcomes. However, as Almond and Currie ( 2011 ) and Almond et al ( 2018 ) point out in their extensive reviews of this wide-ranging literature, more work is needed to disentangle the biological from the more indirect socio-economic mechanisms. Some of examples of these during war could include lack of access to health and policing services, disruption of markets and other key institutions, disturbance of family life, established norms and social networks, and changes to parenting behaviour. We reflect on some of these issues further below when looking at the mechanisms through which exposure to war might affect domestic violence in adulthood.
To test the robustness of our difference-in-differences strategy which assumes parallel trends, we estimate two placebo regressions (using similar methods to for e.g. Akresh et al. 2012a ; Gutierrez and Gallegos 2016 and Weldeegzie 2017 ). In the first test (column 1 of Table 6 ), we exclude the main war-exposed ethnicities (Igbo and other ethnic minorities) and placebo-treat the ethnic groups in the northern part of the country (Kanuri, Hausa, and Fulani), with the remaining ethnicities used as the control group. We choose the northern part of the country given the geographical distance from the area where the war was fought. In the second test (column 2), we placebo-treat the cohort born immediately after the civil war (from 1971 to 1976), with the cohort born from 1977 to 1980 used as the control group. Footnote 7 We would not expect an effect for women born after the civil war. Neither of the coefficients on the placebo-treated interaction term in Table 6 is statistically significant, providing support in favour of our identification strategy. Footnote 8
Although we chose to use the DHS 2008 for this study, as it provides the largest sample of women exposed to the war in childhood (from in utero to age 12), we also check whether our main results hold using the later round of the DHS from 2013. Column 1 of Table 7 shows the estimated effect of war exposure in childhood (without disaggregating across the cohorts) when only the 2013 sample is used, and column 2 of Table 7 shows the estimated effect when the 2008 and 2013 samples are pooled. The results remain robust, with the effect even larger at 5.4 percentage points in column 1 and 4.7 percentage points in column 2 (compared to the 1.2 percentage points estimated in column 4, Table 4 , using the same specification).
In column 3 of Table 7 , we disaggregate the war-exposed women into the four birth cohorts using the pooled sample from 2008 and 2013. Footnote 9 Again, we find the strongest effect from exposure in utero of 5.1 percentage points (compared to 7.4 percentage points in column 4 of Table 5 , using the same specification). However, in the pooled sample, we also find a significant effect of exposure by those exposed between 8 and 12 years. On the whole, though, our robustness checks support our main findings, namely that war exposure in childhood results in a higher incidence of domestic violence among women in adulthood, and that exposure in utero appears to have the strongest effect.
Normalisation of violence.
This section explores two potential mechanisms through which exposure to civil war during childhood may affect the incidence of domestic violence in adulthood. The first is the normalisation of violence hypothesis, which has also been referred to as the intergenerational transmission of violence hypothesis or the model of social learning. Exposure to violence at home during a child’s formative years is known to result in a greater likelihood of being a victim or perpetrator of domestic violence in adulthood (Schwab-Stone et al. 1995 ; Gage 2005 ; Mihalic and Elliott 2007; Yount and Li 2009 ; Cesur and Sabia 2016 ; Jin et al. 2017 ). Along the same lines, one might expect that children exposed to community-level violence during war might also be more likely to view violence as a justifiable response to certain problems (Barnett et al. 2005 ; Fowler et al. 2009 ). In Table 8 , we estimate the effect of women’s exposure to the civil war on the justification of domestic violence to test whether women who were exposed to the conflict in childhood have different attitudes towards domestic violence in adulthood.
Most of the coefficients are positive, many are statistically significant, and some are quite large. In general, the results suggest that, across the birth cohorts, women exposed to the war in childhood are more likely to justify the use of wife-beating than non-exposed women, particularly if the woman argues with her husband, refuses to have sex with him, or burns the food. For example (from row 1), women exposed to war in utero were 2.4 percentage points more likely to justify wife-beating if the woman argues with her husband and 6 percentage points more likely to justify wife-beating if she burns the food, compared to the non-exposed group. The effects are similarly large (and in some cases larger) among those exposed between the ages of 0–4, 5–8, and 9–12, depending on the question asked.
In Table 9 , we use the matched couple’s recode data from the DHS Footnote 10 to investigate the effect of husbands’ exposure to the civil war on the justification of domestic violence in adulthood. This recognises that domestic violence involves both a perpetrator and a victim. Given the high degree of assortative mating by ethnicity in Nigeria, the majority of women who were exposed to the civil war are married to men who were also exposed to the civil war. Indeed, the DHS data indicate that 93.4% of war-exposed women were married to war-exposed men (with only 6.3% of non-exposed women married to war-exposed men). Footnote 11 Because the DHS interviews men aged 15–59, we can disaggregate exposure into in utero, between the ages of 0–4 (born 1966–1970), between the ages of 5–8 (born 1962–1965), between the ages of 9–12 (born 1958–1961), and between the ages of 13–22 (born 1948–1957). The results suggest that compared to non-exposed men, war-exposed men are more likely to justify the use of wife-beating. Although the pattern is not entirely consistent across the five columns, the effect is largest for cohorts of men exposed in utero and between the ages of 9–12 and 13–22.
In addition to being exposed to more community-level violence growing up during war, and marrying men similarly exposed as children, the women exposed to war in childhood may also have been witness to more domestic violence in their own childhood homes or more violent forms of parenting. This could be the case if the stresses and violence of war and the disruption to social norms and family life in turn led to more violence among the parents. The literature summarised in the introduction certainly suggests that intimate partner violence rises during times of war and conflict among married or partnered couples (La Mattina 2017 ; Kelly et al. 2018 ; Østby et al 2019 ; Svallfors 2023 ). The questionnaire asks women if they were aware of domestic violence among their parents, specifically whether the father ever ‘beat’ the mother. We find that 11 percent of women not exposed to the war in childhood were aware of domestic violence among their parents, compared to 19 percent of war-exposed women. This is a substantial and significant difference.
We include this variable as an explanatory variable in the regression and we also interact this variable with the war exposure variables to test whether the effect is stronger for those growing up in the midst of the war. Indeed, in Table 10 , we find a strong positive effect of witnessing domestic violence among one’s parents on the likelihood of becoming a victim oneself in adulthood, and particularly for those exposed to the war in utero. This is a striking result and could suggest that the levels of violence in those war-exposed families where the mother was pregnant were particularly severe, as the combined stresses of war and having another child on the way took their toll. It is also possible that the final months of the war (when these exposed women would have been in utero) were particularly intense, and so the effect on family life more substantial. Finally, disruptions during war to the resources that would ordinarily help mitigate the negative effects of intimate partner violence, such as health and policing services and established social networks, might have exacerbated the experiences of pregnant mothers in particular.
The second mechanism we explore is the intra-household bargaining power hypothesis. Women with limited resources tend to have fewer outside options which can result in an increased likelihood that they will be victims of domestic violence (Gelles 1976 ; Aizer 2010 ). The literature on the effects of conflict provides a number of reasons why women exposed to war may have fewer outside options. Civil conflict results in poorer educational outcomes (Akresh and Walque 2008 ; Leon 2012 ; Shemyakina 2011 ; Chamarbagwala and Moran 2011 ; and Dabalen and Paul 2014 ), and there is evidence that exposure to conflict negatively affects girls more than boys in terms of educational outcomes (Singh and Shemyakina 2016 ). Women with lower education have fewer out-of-marriage options given their weaker labour market outcomes and increased financial dependence on their husbands (Lundberg and Pollak 1996 ; Farmer and Tiefenthaler 1997 ; Aizer 2010 ; Bhattacharyya et al. 2011 ; Eswaran and Malhotra 2011 ; Galdo 2013 ; Heath 2014 ). Furthermore, war exposure can affect marriage, reproductive and health outcomes, which would have consequences for women’s intra-household bargaining power and experiences of domestic violence (Verwimp and van Bavel 2005 ; Akresh 2012a; Grimard and Laszlo 2014 ; Islam et al 2016 ; Cetorelli and Khawaja 2017 ; La Mattina 2017 ).
We test whether war-exposed women have lower bargaining power compared to non-exposed women using the information on decision-making in the household as a proxy. Specifically, we examine whether war-exposed women are less likely to have the final say on certain key decisions in the household compared to non-exposed women. The results in Table 11 show that while most of the coefficients are negative, as predicted, not all are significant. The strongest results are for those exposed in utero; exposure to the civil war decreases the probability of these women having a final say on their own health care by 5.4 percentage points, and on household purchases of daily needs by 8 percentage points. There are also some significant effects, ranging between 3.6 and 5.6 percentage points, for those exposed to the war between the ages of 5–8 and 9–12 for a number of the outcomes.
In this paper, we examine the impact of exposure to war during childhood on women’s experience of domestic violence in adulthood. Unlike other studies that use current geography-based variables to identify exposure to conflict, we are able to use ethnicity and birth cohort given the nature of the Nigerian civil war, thereby mitigating concerns of selective migration. Our results indicate that exposure to the Nigerian civil war during childhood increases the likelihood of women being victims of domestic violence in adulthood, with larger effects for those exposed at younger ages, and particularly large effects for those exposed in utero. This is consistent with evidence to suggest that the early childhood period, including the time in utero, is particularly important for later life outcomes and that shocks during this period can have long-lasting effects.
Understanding the mechanisms through which civil war affects domestic violence is equally as important as identifying the effect itself, especially if effective post-war policies are to be designed to mitigate the deleterious consequences of conflict in developing countries. However, identifying the mechanisms is a much more difficult task with the data available, and therefore, our results can only be interpreted as suggestive.
First, we find that both the women in our sample and their husbands who were exposed to the war during childhood are more likely to perceive domestic violence to be an acceptable behaviour in adulthood than those not exposed to the war. This is in line with the normalisation of violence hypothesis that predicts that those exposed to violence in childhood are more likely to become either perpetrators or victims of domestic violence in adulthood. In addition, we find war-exposed women were more likely to witness domestic violence in their own childhood homes than non-exposed women, and that witnessing domestic violence among their parents is positively correlated with experiencing domestic violence themselves in adulthood particularly among those exposed in utero. It is possible that the combined stresses of war and having another child on the way led to more violent behaviour in the home, or that the final months of war (when these exposed women would have been in utero) were particularly intense, and so the effect on family life more marked. Footnote 12
Second, our findings suggest that women who were exposed to the war in childhood also have lower intra-household bargaining power compared to non-exposed women, which would make them more vulnerable to incidents of domestic violence. Relative to the non-exposed group, we found women who were exposed to the conflict in childhood have less decision-making power in their households in adulthood, and again the effect appears stronger among those in utero (although there is evidence also for the other cohorts). This might be the case if war exposure affected women’s educational, health, and reproductive outcomes in ways that placed them in a more precarious position relative to men in the marriage market.
However, this is a subject for further study given the complexity of the potential pathways and mechanisms. The large effects measured for children who were exposed to the war in utero in particular warrant further investigation. These results are consistent with the evidence from a large literature showing that conditions and events in utero can have long-lasting consequences for the individual’s physical and mental health as well as their education, employment, and earnings outcomes (Ravelli et al 1976 ; Hoek et al 1998 ; Almond 2006 ; Dunn 2007 ; Camacho 2009 ; Almond and Currie 2011 ; Comfort 2016 ). However, much more work is needed to disentangle the biological from the social mechanisms in order to better understand both the direct and more indirect channels through which foetal-origin effects are generated (Almond and Currie 2011 ; Almond et al. 2018 ).
The relevance of our study and the need for further work in this area is underscored by the pervasiveness of domestic violence. A recent study estimated the global prevalence of intimate partner violence to be around 30%, and for the sub-Saharan African region specifically, closer to 37% (WHO 2017 ). Moreover, the consequences of domestic violence, both human and economic, are substantial. Domestic violence results in direct physical and mental harm to women, with research pointing to poorer health outcomes and a greater likelihood of depressive symptoms and substance abuse among victims (Coker et al. 2002 ; Silverman et al. 2006 ; Ackerson et al. 2008 ; Ellsberg et al. 2008 ; Meekers et al. 2013 ). Domestic violence can also result in substantial economic costs related to policing, health expenditure, and reduced economic productivity (Walby 2004 ). Lastly, children of women who experience domestic violence have worse outcomes, such as lower birth weight, lower IQ scores, a greater likelihood of emotional and behavioural problems, and a higher probability of acquiring HIV (Sternberg et al. 1993 ; Koenen et al. 2003 ; Aizer 2011 ; WHO 2013 ; Rawlings and Siddique 2014 , 2018 ; Currie et al 2022 ). Understanding both the causes and longer-term implications of domestic violence is imperative to designing appropriate policy responses and support mechanisms.
The dataset used to obtain the results for this paper can be made available upon request.
These three main regions were subsequently demarcated into six geopolitical regions, namely the northeast, northwest, north-central, south-south, south-east, and south-west, the latter being the region where the civil war was fought (Alapiki 2005 ). These six regions are further divided into 36 states.
The 2008 Nigerian Demographic Health survey also interviewed men aged 15 to 59 to provide information on health and other related issues, but it did not collect information on their experiences of domestic violence.
We were unable to analyse exposure after age 12 (or among cohorts born pre-1958) because the DHS contains information only on women aged 15 to 49 years old. In the 2008 DHS wave, the oldest woman in the sample (aged 49) therefore was born in 1958. If we use later waves of the DHS, we can only analyse a smaller sample of war-exposed women. Specifically, if we used the 2013 DHS, we would only be able to estimate the effect for those exposed in utero to age 7, and if we used the 2018 DHS, we would only be able to estimate the effect for those exposed in utero to age 2.
The DHS captures information on experiences of domestic violence using the World Health Organization’s ethical and safety guidelines (Kishor and Kiersten 2004 ). Interviewers are trained to deal with the sensitive nature of the questions and there are strict protocols to ensure privacy during the interview. To try to minimise under-reporting of domestic violence, the DHS domestic violence questionnaire uses a modified version of the Conflict Tactics Scale (CTS). Women are asked a number of separate questions on different types of violence which reduces confusion as to what constitutes domestic violence, and gives women multiple opportunities to reveal their experiences (Kishor 2005 ).
We limit our control group to the six-year period following the war, as too broad a window of comparison increases potential confounding effects (Akresh et al 2012a ). Moreover, our results are consistent when, following Akresh et al ( 2012a ), we use an even shorter control period, namely 1970 (Nov) to 1974.
If the immediate post-war environment in the south-eastern region did not experience a full recovery, then these impacts of war exposure would be underestimated, and our findings would represent a lower-bound effect.
To validate the placebo result, we conducted further robustness checks using equal intervals of years for the treatment and control groups (1971–1974 and 1975–1978). We find statistically insignificant effects of exposure to civil war on domestic violence in these additional checks.
Akresh et al ( 2012a ) run slightly different placebo tests on ethnic group and cohort but similarly find no significant effects. They also use estimated ethnic mortality during the war instead of ethnicity itself in their regressions to test for the validity of the identification strategy and find remarkably similar results. This leads them to conclude that the strategy to use ethnicity to identify exposure “while simple, is accurate and powerful” (Akresh et al. 2012a : 275).
Because the DHS only interviews women aged 15 to 49, the oldest women included in the 2013 survey would have been born in 1964, and therefore, we can only capture war exposure from in utero through to age 7. To estimate the exposure by birth cohort, we therefore only show the results using the pooled 2008 and 2013 datasets. We did not attempt to include the 2018 DHS in the robustness checks, as the sample of war-exposed women would have shrunk even further to those women who were exposed in utero through to 2 years of age.
The DHS couple’s recode data contain information on the husbands/partners (aged 15–59) for the sample of women who were married/cohabiting and living with their partners during the interview.
The high level of intra-ethnic marriage is consistent with low levels of migration across states, with most migration in Nigeria occurring within states from rural to urban areas (Federal Office of Statistics 1999 ; 2000).
Unfortunately, we are unable to test more formally for a relationship between the intensity of conflict and domestic violence. To do so would require data on the variation in the number of deaths caused by the civil war across districts and time, and to the best of our knowledge, no such data exist (there are only estimates of the total number of deaths caused by the war).
Aall, C. 1970. Relief, Nutrition and Health in the Nigerian/Biafran War. Journal of Tropical Pediatrics 16 (2): 70–90.
Article Google Scholar
Ackerson, L.K., I. Kawachi, E.M. Barbeau, and S.V. Subramanian. 2008. Effects of Individual and Proximate Educational Context on Intimate Partner Violence: A Population-Based Study of Women in India. American Journal of Public Health 98 (3): 507–514.
Aizer, A. 2010. The Gender Wage Gap and Domestic Violence. American Economic Review 100 (4): 1847–1859.
Aizer, A. 2011. Poverty, Violence, and Health: The Impact of Domestic Violence During Pregnancy on Newborn Health. Journal of Human Resources 46 (3): 518–538. https://doi.org/10.1353/jhr.2011.0024 .
Ajefu, J.B., and D. Casale. 2021. The Long-Term Effects of Violent Conflict on Women’s Intra-household Decision-Making Power. Journal of Development Studies 57 (10): 1690–1709. https://doi.org/10.1080/00220388.2021.1873285 .
Akbulut-Yuksel, M. 2017. War During childhood: The Long Run Effects of Warfare on Health. Journal of Health Economics 53: 117–130.
Akresh, R., S. Bhalotra, M. Leone, and U.O. Osili. 2012a. War and Stature: Growing Up During the Nigerian Civil War. American Economic Review 102 (3): 273–277.
Akresh, R., L. Lucchetti, and H. Thirumurthy. 2012b. Wars and Child Health: Evidence from the Eritrean-Ethiopian Conflict. Journal of Development Economics 99: 330–340.
Akresh, R. and de Walque, D. 2008. Armed Conflict and Schooling: Evidence from the 1994 Rwandan Genocide, The World Bank, Washington, D.C., Policy Research Working Paper No. 4606.
Akresh, R., S. Bhalotra, M. Leone, and U. Osili. 2023. First-and Second-Generation Impacts of the Biafran War. Journal of Human Resources 58 (2): 488–531.
Alapiki, H.E. 2005. State Creation in Nigeria: Failed Approached to National Integration and Local Autonomy. African Studies Review 48 (3): 49–65.
Almond, D. 2006. Is the 1918 Influenza Pandemic Over? Long-Term Effects of In Utero Influenza Exposure in the Post-1940 U.S. Population. Journal of Political Economy 14 (4): 672–712.
Almond, D., and J. Currie. 2011. Killing me Softly: The Fetal Origins Hypothesis. The Journal of Economic Perspectives 25 (3): 153–172.
Almond, D., J. Currie, and V. Duque. 2018. Childhood Circumstances and Adult Outcomes: Act II. Journal of Economic Literature 56 (4): 1360–1446.
Barker, D.J. 1990. The Fetal and Infant Origins of Adult Disease. BMJ 301 (6761): 1111.
Barker, D.J. 1995. Fetal Origins of Coronary Heart Disease. BMJ 311 (6998): 171–174.
Bertrand, M., E. Duflo, and S. Mullainathan. 2004. How Much Should We Trust Differences-in-Differences Estimates? The Quarterly Journal of Economics 119 (1): 249–275.
Barnett, O., C.L. Miller-Perrin, and R.D. Perrin. 2005. Family Violence Across the Lifespan: An introduction . Thousand Oaks, CA: Sage.
Google Scholar
Bhattacharyya, M., A.S. Bedi, and A. Chhachhi. 2011. Marital Violence and Women’s Employment and Property Status: Evidence from North Indian Villages. World Development 39 (9): 1676–1689.
Camacho, A. 2009. Stress and Birth Weight: Evidence from Terrorist Attacks. American Economic Review: Papers & Proceedings. 98 (2): 511–515.
Case, A., A. Fertig, and C. Paxson. 2005. The Lasting Impact of Childhood Health and Circumstance. Journal of Health Economics 24: 365–389.
Cesur, R., and J.J. Sabia. 2016. When War Comes Home: The Effect of Combat Service on Domestic Violence. Review of Economics and Statistics 98 (2): 209–225.
Cetorelli, V., and Khawaja. 2017. Intensity of Conflict and Fertility in the Occupied Palestian Territory: A Longitudinal Study. The Lancet 390 (2): 350.
Chamarbagwala, R., and H. Moran. 2011. The Human Capital Consequences of Civil War: Evidence from Guatemala. Journal of Development Economics 94: 41–61.
Coker, A.L., K.E. Davis, I. Arias, S. Desai, M. Sanderson, H.M. Brandt, and P.H. Smith. 2002. Physical and Mental Health Effects of Intimate Partner Violence for Men and Women. American Journal of Preventive Medicine 23 (4): 260–268.
Comfort, A.B. 2016. Long-Term Effect of In Utero Conditions on Maternal Survival Later in Life: Evidence from Sub-Saharan Africa. Journal of Population Economics. 29 (2): 493–527.
Cunha, F., and J. Heckman. 2007. The Technology of Skill Formation. American Economic Review 97 (2): 31–47.
Currie, J. 2020. Child Health as Human Capital. Health Economics. 29: 452–463. https://doi.org/10.1002/hec.3995 .
Currie, J., M. Mueller-Smith, and M. Rossin-Slater. 2022. Violence While in Utero: The Impact of Assaults During Pregnancy on Birth Outcomes. The Review of Economics and Statistics 104 (3): 525–540. https://doi.org/10.1162/rest_a_00965 .
Dabalen, A.L., and S. Paul. 2014. Estimating the Effects of Conflict on Education in Cote d’Ivoire. The Journal of Development Studies 50 (12): 1631–1646.
Dunn, P.M. 2007. Perinatal Lessons from the Past: Sir Norman Gregg, ChM, MC, of Sydney (1892–1966) and Rubella Embryopathy. Archives of Disease in Childhood 92 (6): F513–F514. https://doi.org/10.1136/adc.2005.091405 .
Ellsberg, M., H.A.F.M. Jansen, et al. 2008. Intimate partner Violence and Women’s Physical and Mental Health in the WHO multi-country study on Women’s Health and Domestic Violence: An Observational Study. The Lancet 371 (9619): 1165–1172.
Eswaran, M., and N. Malhotra. 2011. Domestic Violence and Women’s Autonomy in Developing Countries: Theory and Evidence. Canadian Journal of Economics. 44 (4): 1222–1263.
Farmer, A., and J. Tiefenthaler. 1997. An Economic Analysis of Domestic Violence. Review of Social Economy 55 (3): 337–358.
Federal Office of Statistic. 1999. Annual Abstract of Statistics Various Years.
Federal Office of Statistics. 2000 Social Statistics in Nigeria Various Years.
Fowler, P.J., C.J. Tompsett, Jordan M. Braciszewski, Angela J. Jaques-Tiura, and B.B. Baltes. 2009. Community Violence: A Meta-Analysis on the Effect of Exposure and Mental Health Outcomes of Children and Adolescents. Development and Psychopathology 21: 227–259.
Gage, A. 2005. Women’s Experience of Intimate Partner Violence in Haiti. Social Science and Medicine 61: 343–364.
Galdo, J. 2013. The Long-Run Labor-Market Consequences of Civil War: Evidence from the Shining Path in Peru. Economic Development and Cultural Change 61 (4): 789–823.
Gelles, R.J. 1976. Abused Wives: Why Do They Stay? Journal of Marriage and the Family 38 (4): 659–667.
Gleditsch, N.P., P. Wallensteen, M. Eriksson, M. Sollenberg, and H. Strand. 2002. Armed Conflict 1946–2001: A New Dataset. Journal of Peace Research 39 (5): 615–637.
Grimard, F., and S. Laszlo. 2014. Long-Term Effects of Civil Conflict on Women’s Health Outcomes in Peru. World Development 54: 139–155.
Gutierrez, I.A. and Gallegos, J.V. 2016. The Effect of Civil War on Domestic Violence: The Case of Peru, Working Paper, RAND Labour and Population, WR-1168.
Heath, R. 2014. Women’s Access to Labour Market Opportunities, Control of Household Resources, and Domestic Violence: Evidence from Bangladesh. World Development 57: 32–46.
Hoek, H.W., A.S. Brown, and E. Susser. 1998. The Dutch Famine and Schizophrenia Spectrum Disorders. Social Psychiatry and Psychiatric Epidemiology 33 (8): 373–379.
Huber, Laura. 2023. One Step Forward, One Step Back: The Micro-Level Impacts of Conflict on Women’s Security. International Studies Quarterly 67 (2): 019.
Institute for Economics and Peace. 2018. The Economic Value of Peace 2018: Measuring the Global Economic Impact of Violence and Conflict, Sydney.
Islam, A., C. Ouch, R. Smyth, and L.C. Wang. 2016. The Long-term effects of Civil Conflicts on Education, Earnings, and Fertility: Evidence from Cambodia. Journal of Comparative Economics 44: 800–820.
Jin, X., T. Yang, and M.W. Feldman. 2017. Intergenerational Transmission of Marital Violence among Rural Migrants in China: Evidence from a Survey in Shenzhen. Journal of Contemporary China 26 (108): 931–947.
Kelly, J.T.D., E. Colantuoni, C. Robinson, and M.R. Decker. 2018. From the Battlefield to the Bedroom: A Multilevel Analysis of the Links Between Political Conflict and Intimate Partner Violence in Liberia. BMJ Global Health 3: e000668. https://doi.org/10.1136/bmjgh-2017-000668 .
Kijewski, S., and M. Freitag. 2018. Civil War and the Formation of Social Trust in Kosovo: Post-traumatic Growth or War-related Distress? Journal of Conflict Resolution 62 (4): 717–742.
Kirk-Greene, A.H.M. 1971. Crisis and Conflict in Nigeria . Oxford: Oxford University Press.
Kishor, S., and J. Kiersten. 2004. Profiling Domestic Violence: A Multi-country Study . Calverton, Maryland: ORC Macro.
Kishor, S. 2005. Domestic Violence Measurement in the Demographic and Health Surveys: The History and the Challenges, Paper Presented at Expert Group Meeting.
Koenen, K.C., T.E. Moffitt, et al. 2003. Domestic Violence is Associated with Environmental Suppression of IQ in Young Children. Development and Psychopathology 15 (02): 297–311.
La Mattina, G. 2017. Civil Conflict, Domestic Violence and Intra-household Bargaining in Post-Genocide Rwanda. Journal of Development Economics 124: 168–198.
La Mattina, G. and Shemyakina, O.N. 2017. Domestic Violence and Childhood Exposure to Armed Conflict: Attitudes and Experiences, Unpublished Manuscript.
Leon, G. 2012. Civil Conflict and Human Capital Accumulation: The Long-Term Effects of Political Violence in Peru. Journal of Human Resources 47 (4): 992–1022.
Lundberg, S., and R.A. Pollak. 1996. Bargaining and Distribution in Marriage. The Journal of Economic Perspectives 10 (4): 988–158.
Mihalic, S.W., and D. Elliott. 1997. A Social Learning Theory Model of Marital Violence. Journal of Family Violence 12 (1): 21–47.
Meekers, D., S. Pallin, and P. Hutchinson. 2013. Intimate Partner Violence and Mental Health in Bolivia. BMC Women’s Health 13: 1.
Nafziger, E.W. 1972. The Economic Impact of the Nigerian Civil War. The Journal of Modern African Studies 10 (2): 223–245.
Østby, Gudrun, Michele Leiby, and Ragnhild Nordås. 2019. The Legacy of Wartime Violence on Intimate-Partner Abuse: Microlevel Evidence from Peru, 1980–2009. International Studies Quarterly 63 (1): 1–46.
Ravelli, G.P., Z.A. Stein, and M.W. Susser. 1976. Obesity in Young Men After Famine Exposure In Utero and Early Infancy. New England Journal of Medicine 295 (7): 349–353.
Rawlings, S. and Siddique, Z. 2014. Domestic Abuse and Child Health, IZA Discussion Paper No. 8566.
Rawlings, S. and Siddique, Z. 2018. Domestic Violence and Child Mortality, IZA Discussion Paper No. 11899.
Schwab-Stone, M.E., Tim S. Ayers, K. Wesley, V. Charlene, B. Charles, S. Timothy, and R.P. Weissberg. 1995. No Safe Haven: A Study of Violence Exposure in an Urban Community. Journal of the American Academy of Child and Adolescent Psychiatry 34: 1343–1352.
Shemyakina, O. 2011. The Effect of Armed Conflict on Accumulation of Schooling: Results from Tajikistan. Journal of Development Economics 95 (2): 186–200.
Silverman, J., M. Decker, E. Reed, and A. Raj. 2006. Intimate Partner Violence Victimization Prior To and During Pregnancy Among Women Residing in the 26 United States: Associations with Maternal and Neonatal Health. American Journal of Obstetrics and Gynecology 195 (1): 140–148.
Simpson, B. 2014. The Biafran Secession and the Limits of Self-Determination. Journal of Genocide Research 16 (2–3): 337–354.
Singh, P., and O.N. Shemyakina. 2016. Gender-Differential Effects of Terrorism on Education: The Case of the 1981–1993 Punjab Insurgency. Economics of Education 54: 185–210.
Sternberg, K.J., M.E. Lamb, C. Greenbaum, D. Cicchetti, D. Samia, R.M. Cortes, O. Krispin, and F. Lorey. 1993. Effects of Domestic Violence on Children’s Behaviour Problems and Depression. Developmental Psychology 29 (1): 44–52.
Svallfors, S. 2023. Hidden Casualties: The Links Between Armed Conflict and Intimate Partner Violence in Colombia. Politics & Gender 1–33: 1–33. https://doi.org/10.1017/S1743923X2100043X .
Swee, E.L. 2015. On War Intensity and Schooling Attainment: The Case of Bosnia and Herzegovina. European Journal of Political Economy 40: 158–172.
Torrisi, O. 2023. Young-Age Exposure to Armed Conflict and Women’s Experiences of Intimate Partner Violence. Journal of Marriage and Family 85 (1): 7–32.
Udo, R.K. 1970. Reconstruction in the War-Affected Areas of Nigeria. The Royal Geographical Society 2 (3): 9–12.
Verwimp, P., and J. van Bavel. 2005. Child Survival and Fertility of Refugees in Rwanda. European Journal of Population 21 (2): 271–290.
Verwimp, P., P. Justino, and T. Brück. 2019. The Microeconomics of Violent Conflict. Journal of Development Economics 141: 102297.
Volpe, E.M., T.L. Hardie, C. Cerulli, M.S. Sommers, and D. Morrison-Beedy. 2013. What’s Age Got to Do with It? Partner Age Difference, Power, Intimate Partner Violence, and Sexual Risk in Urban Adolescents. Journal of Interpersonal Violence 28 (10): 2068–2087.
Walby, S. 2004. The Cost of Domestic Violence, Women and Equality Unit (DTI).
Weldeegzie, S.G. 2017. Growing-up Unfortunate: War and Human Capital in Ethiopia. World Development 96: 474–489.
World Health Organization. 2013. Global and Regional Estimates of Violence against Women, Prevalence and Health Effects of Intimate Partner Violence and Non-Partner Sexual Violence, World Health Organisation.
World Health Organization. 2014. Health Care for Women Subjected to Intimate Partner Violence or Sexual Violence . A Clinical Handbook Report: World Health Organization.
World Health Organization. 2017. Global and Regional Estimates of Violence against Women: Prevalence and Health Effects of Intimate Partner Violence and Non-Partner Sexual Violence, World Health Organization.
World Development Report. 2011. Conflict, Security, and Development . Washington DC: The World Bank.
World Humanitarian Data and Trends Report. 2017. A Report from the UN Office for the Coordination of Humanitarian Affairs.
Yount, K., and L. Li. 2009. Women’s “Justification” of Domestic Violence in Egypt. Journal of Marriage and Family 71 (5): 1125–1140.
Download references
Authors and affiliations.
Department of Peace Studies and International Development, Faculty of Management, Law, and Social Sciences, University of Bradford, Bradford, UK
Joseph B. Ajefu
Centre for Social Development in Africa (CSDA), University of Johannesburg, Johannesburg, South Africa
School of Economics and Finance, University of the Witwatersrand, Johannesburg, South Africa
Daniela Casale
You can also search for this author in PubMed Google Scholar
Correspondence to Joseph B. Ajefu .
Publisher's note.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .
Reprints and permissions
Ajefu, J.B., Casale, D. Long-Term Effects of Childhood Exposure to War on Domestic Violence. Eur J Dev Res (2024). https://doi.org/10.1057/s41287-024-00659-4
Download citation
Accepted : 14 July 2024
Published : 22 August 2024
DOI : https://doi.org/10.1057/s41287-024-00659-4
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative
IMAGES
VIDEO
COMMENTS
Statistical Treatment Example - Quantitative Research. For a statistical treatment of data example, consider a medical study that is investigating the effect of a drug on the human population. As the drug can affect different people in different ways based on parameters such as gender, age and race, the researchers would want to group the ...
Research Paper Statistical Treatment of Data: A Primer. March 11, 2024. We can all agree that analyzing and presenting data effectively in a research paper is critical, yet often challenging. This primer on statistical treatment of data will equip you with the key concepts and procedures to accurately analyze and clearly convey research findings.
This article is a practical introduction to statistical analysis for students and researchers. We'll walk you through the steps using two research examples. The first investigates a potential cause-and-effect relationship, while the second investigates a potential correlation between variables. Example: Causal research question.
Introduction. Statistical analysis is necessary for any research project seeking to make quantitative conclusions. The following is a primer for research-based statistical analysis. It is intended to be a high-level overview of appropriate statistical testing, while not diving too deep into any specific methodology.
Type and distribution of the data used. For the same objective, selection of the statistical test is varying as per data types. For the nominal, ordinal, discrete data, we use nonparametric methods while for continuous data, parametric methods as well as nonparametric methods are used.[] For example, in the regression analysis, when our outcome variable is categorical, logistic regression ...
Abstract. Statistical methods involved in carrying out a study include planning, designing, collecting data, analysing, drawing meaningful interpretation and reporting of the research findings. The statistical analysis gives meaning to the meaningless numbers, thereby breathing life into a lifeless data. The results and inferences are precise ...
ANOVA and MANOVA tests are used when comparing the means of more than two groups (e.g., the average heights of children, teenagers, and adults). Predictor variable. Outcome variable. Research question example. Paired t-test. Categorical. 1 predictor. Quantitative. groups come from the same population.
Treatments are divided into two groups: descriptive statistics, which summarize your data as a graph or summary statistic and inferential statistics, which make predictions and test hypotheses about your data. Treatments could include: Finding T-Scores or Z-Scores. Calculating Correlation coefficients. 2.
A key characteristic of scientific research is that the entire experiment (or series of experiments), including the data analyses, is reproducible. This aspect of science is increasingly emphasized. The Materials and Methods section of a scientific paper typically contains the necessary information for the research to be replicated and expanded on by other scientists. Important components are ...
The new guidelines discuss many aspects of the reporting of studies in the Journal, including a requirement to replace P values with estimates of effects or association and 95% confidence ...
Some of these methods and tools are used within specific fields of research, such as statistical tests of significance, double-blind trials, and proper phrasing of questions on surveys. Others apply across all research fields, such as describing to others what one has done so that research data and results can be verified and extended.
Now, if we divide the frequency with which a given mean was obtained by the total number of sample means (36), we obtain the probability of selecting that mean (last column in Table 10.5). Thus, eight different samples of n = 2 would yield a mean equal to 3.0. The probability of selecting that mean is 8/36 = 0.222.
A bstract. The choice of statistical test used for analysis of data from a research study is crucial in interpreting the results of the study. This article gives an overview of the various factors that determine the selection of a statistical test and lists some statistical testsused in common practice. How to cite this article: Ranganathan P.
Furthermore, statistics in research helps interpret the data clustered near the mean of distributed data or spread across the distribution. These trends help analyze the sample and signify the hypothesis. 3. Data Interpretation Through Analysis. When dealing with large data, statistics in research assist in data analysis. This helps researchers ...
Statistical treatment of data also involves describing the data. The best way to do this is through the measures of central tendencies like mean, median and mode. These help the researcher explain in short how the data are concentrated. Range, uncertainty and standard deviation help to understand the distribution of the data.
Types of descriptive statistics. There are 3 main types of descriptive statistics: The distribution concerns the frequency of each value. The central tendency concerns the averages of the values. The variability or dispersion concerns how spread out the values are. You can apply these to assess only one variable at a time, in univariate ...
Statistical methods appropriate in research are described with examples. Topics covered include the choice of appropriate averages and measures of dispersion to summarize data sets, and the choice of tests of significance, including t-tests and a one- and a two-way ANOVA plus post-tests for normally distributed (Gaussian) data and their non-parametric equivalents.
Mean - the arithmetic average, calculated by summing all the values and dividing by the number of values in the sum. Median - the mid point of the distribution, where half the values are higher and half lower. Mode - the most frequently occurring value. Range - the difference between the highest and lowest value.
Correlation analysis has three general outcomes: (1) the two variables rise and fall together; (2) as values in one variable rise, the other falls; and (3) the two variables do not appear to be systematically related. To make those determinations, we use the correlation coefficient (r) and related p value or CI.
Research Design and Methodology. Chapter 3 consists of three parts: (1) Purpose of the. study and research design, (2) Methods, and (3) Statistical. Data analysis procedure. Part one, Purpose of ...
The methods section should describe what was done to answer the research question, describe how it was done, justify the experimental design, and explain how the results were analyzed. Scientific writing is direct and orderly. Therefore, the methods section structure should: describe the materials used in the study, explain how the materials ...
Inferential statistics. In inferential statistics, data from a sample is analysed to conclude the entire population. The goal is to prove or disprove the theories. A hypothesis is a suggested explanation for a phenomenon (plural hypotheses). Hypothesis testing is essential to process for making logical choices regarding observed effects ...
Researchers must utilize exploratory data techniques to present findings to a target audience and create appropriate graphs and figures. Researchers can determine if outliers exist, data are missing, and statistical assumptions will be upheld by understanding data. Additionally, it is essential to comprehend these data when describing them in conclusions of a paper, in a meeting with ...
This paper highlights the scarring effects of early life exposure to civil war, by examining the impact of exposure to conflict in childhood on the incidence of domestic violence in adulthood among married women. To estimate these effects, we use a difference-in-differences model which exploits variation in exposure to Nigeria's 30-month-long civil war by year of birth and ethnicity. Our ...