The Household Survey

265 Pages·2017·3.5 MB·English

by  Neil Stewart

City Research Online
City, University of London Institutional Repository
Citation: Andersson, Neil (2013). Uncertainties in gender violence epidemiology.
(Unpublished Doctoral thesis, City University London)
This is the unspecified version of the paper.
This version of the publication may differ from the final published version.
Permanent repository link: https://openaccess.city.ac.uk/id/eprint/2420/
Link to published version:
Copyright: City Research Online aims to make research outputs of City,
University of London available to a wider audience. Copyright and Moral Rights
remain with the author(s) and/or copyright holders. URLs from City Research
Online may be freely distributed and linked to.
Reuse: Copies of full items can be used for personal research or study,
educational, or not-for-profit purposes without prior permission or charge.
Provided that the authors, title and full bibliographic details are credited, a
hyperlink and/or URL is given for the original metadata page and the content is
not changed in any way.
City Research Online: http://openaccess.city.ac.uk/ [email protected]
Uncertainties in gender violence epidemiology
Neil Andersson
Thesis submitted for the degree of
Doctor of Philosophy (PhD) by prior publication
City University London
School of Health Sciences
April 2013
TABLE OF CONTENTS
List of Tables .................................................................................................................5(cid:3)
List of Figures................................................................................................................6(cid:3)
List of Appendices ........................................................................................................7 (cid:3)
Abbreviations ................................................................................................................8 (cid:3)
Glossary .........................................................................................................................9(cid:3)
Acknowledgements .....................................................................................................1 3(cid:3)
Declaration ...................................................................................................................14(cid:3)
ABSTRACT...................................................................................................................1 5(cid:3)
CHAPTER 1. INTRODUCTION ....................................................................................1 6(cid:3)
Summary ..................................................................................................................1 6(cid:3)
1.1 CIET and the research programme ....................................................................1 6(cid:3)
1.2 Defining terms ....................................................................................................1 8(cid:3)
1.2.1 Gender ........................................................................................................1 8(cid:3)
1.2.2 Violence ......................................................................................................1 9(cid:3)
1.2.3 Gender violence epidemiology ....................................................................2 1(cid:3)
1.3 The papers in this PhD by prior publication ........................................................2 4(cid:3)
1.3.1 Methods in gender violence epidemiology ..................................................2 6(cid:3)
1.3.2 Gender violence and HIV in southern Africa ...............................................2 7(cid:3)
1.3.3 Gender violence and maternal morbidity ....................................................3 1(cid:3)
1.4 Cumulative knowledge gained ............................................................................3 2(cid:3)
1.5 Research question ..............................................................................................3 4(cid:3)
1.5.1 Missing data ................................................................................................3 4(cid:3)
1.5.2 Clustering ....................................................................................................3 4(cid:3)
1.5.3 Unrecognised causality ...............................................................................3 5(cid:3)

2
CHAPTER 2. METHODS ..............................................................................................3 6(cid:3)
Summary ..................................................................................................................3 6(cid:3)
2.1 Critical review of the published papers ...............................................................3 6(cid:3)
2.2 Review of missing data in intimate partner violence studies .............................. 37(cid:3)
2.3 Qualitative research on missing data ..................................................................3 8(cid:3)
2.4 Correlated group data: incidental and informative clustering .............................. 40(cid:3)
2.5 Revisiting gender violence as a covariate ..........................................................4 1(cid:3)
CHAPTER 3. MISSING DATA ......................................................................................4 3(cid:3)
Summary ..................................................................................................................4 3(cid:3)
3.1 Ignorability of missing data .................................................................................4 3(cid:3)
3.2 Review of missing data in IPV studies ................................................................4 6(cid:3)
3.3 Focus group discussions ....................................................................................4 6(cid:3)
3.3.1 Gender roles and “indiscipline” ...................................................................4 7(cid:3)
3.3.2 Non-availability and violence ......................................................................4 8(cid:3)
3.3.3 Fear of violence ..........................................................................................4 8(cid:3)
3.3.4 Experience of violence among non-responders ..........................................5 0(cid:3)
3.4 Adjustments to estimates of rates of violence .....................................................5 0(cid:3)
3.5 An approach to minimise missing data and non-disclosure ................................5 1(cid:3)
3.6 Discussion ..........................................................................................................5 4(cid:3)
CHAPTER 4. CLUSTERING .........................................................................................5 7(cid:3)
Summary ..................................................................................................................5 7(cid:3)
4.1 Informative and incidental clustering ..................................................................5 7(cid:3)
4.2 Reanalysis of clustering for IPV ..........................................................................5 9(cid:3)
4.3 Discussion ..........................................................................................................6 3(cid:3)
CHAPTER 5. GENDER VIOLENCE AND UNRECOGNISED CAUSALITY .................66(cid:3)
Summary ..................................................................................................................6 6(cid:3)

3
5.1 Introduction .........................................................................................................6 6 
5.2 Causes, confounders and colliders.....................................................................6 7 
5.3 Revisiting gender violence as a covariate ..........................................................7 0 
5.4 Discussion ..........................................................................................................7 4 
CHAPTER 6. CONCLUSIONS .....................................................................................7 7 
Summary ..................................................................................................................7 7 
6.1 Less uncertainty in gender violence epidemiology .............................................7 7 
6.1.1 Missing data ................................................................................................7 8 
6.1.2 Clustering ....................................................................................................7 8 
6.1.3 Unrecognised causality ...............................................................................7 9 
6.2 Implications for future research ..........................................................................7 9 
6.3 Endnote ..............................................................................................................8 2 
Paper 1………………………………………….……………………………………………..85
Paper 2………………………………………….……………………………………………..99
Paper 3…………………………………………..………………………………………...…116
Paper 4…………………………………………..……………………………………..….…139
Paper 5…………………………………………..………………………………………...…146
Paper 6…………………………………………..………………………………………...…154
Paper 7…………………………………………..……………………………………..….…166
Paper 8…………………………………………...………………………………………..…181
Paper 9…………………………………………...………………………………………..…197
Paper 10………………………………………...……………………………………………202
Paper 11………………………………………...……………………………………………214
REFERENCES ............................................................................................................2 27
APPENDICES……………………………………………………………………………….234 

4
List of Tables

Table 1. Domestic violence in eight southern African countries, female p60
responses and male group variables as possible covariates of intimate
partner violence (2002)

Table 2. Domestic violence in Zambia, female responses and group variables p61
as covariates of intimate partner violence (2002)

Table 3. The influence of gender violence history in HIV research p72

5
List of Figures

Figure 1. Covariate relationships p68

6
List of Appendices

Appendix 1. 75 studies of intimate partner violence published between 2000 and 2011

Appendix 2. Missing data in 75 studies of intimate partner violence published between
2000 and 2011

Appendix 3. Risk factors and analysis in 75 studies of intimate partner violence
published between 2000 and 2011

7
Abbreviations
GBV - Gender based Violence:
GEE - Generalised Estimating Equation
GLMM - Generalised Linear Mixed Model
IPV – Intimate Partner Violence
MH - Mantel Haenszel
MVA - Multivariate Analysis

8
Glossary
Bias: a systematic favouritism in measurement, usually at the expense of accuracy; a
biased estimate is systematically different from the population parameter of interest.
Collider bias: An association between the base set changes (usually blocked) when
both exposure and outcome are related a third variable in a derived set. Called a
collider because of the pictorial depiction of two causes A and B colliding on C:
A(cid:314)C(cid:312)B
Confounder: a variable extraneous (outside the causal chain) to an association
between exposure and outcome, that is itself associated with both exposure and
outcome, and distorts the measured association between the two.
Constraints: restrictions, like the need to use face to face interviews, limit the meaning
one can draw from gender violence epidemiology; missing data also limit the meaning.
Domestic violence/abuse: also known as family violence, includes intimate partner
violence, child abuse, and elder abuse; can include physical or mental abuse,
controlling behaviour or economic deprivation.
Epidemiology: Presumed to refer to the patterns, causes and effects of health events,
Epi- (over, among) demos- (people) logia (sayings, utterances) refers to occurrence
relations – event patterns in society. It is highly relevant to health and other social
events, but also to environment, animal health, and other fields of study.
Gender: Social defined discriminating characteristics and sexual roles of men and
women.
Gender based violence: Many sources equate GBV with violence against women.
The term includes some kinds of violence against men and male children. In this thesis,

9