Situation of Intimate Partner Violence - A Statistical Study

Arindam Gupta¹, Chowdhury Sadeka Nazrin¹, Sudhansu S. Maiti²

¹Department of Statistics, The University of Burdwan, Burdwan

²Department of Statistics, Visva-Bharati University, Santiniketan-731235, India

*Corresponding Author E-mail: guptaarin@gmail.com

ABSTRACT:

There are many types of violence that could be make human's life miserable. On those types of violence one of the most painful is “Intimate Partner Violence” (IPV). It could be happened with both men and women. In this type of violence intimate partner becomes violent and perform violent acts against their associates. IPV also happened in same sex partnerships. We are trying to compare the IPV situation among the different states in India as well as between two countries “India” and “Bolivia” by using statistical Quality Control chart technique (SQC chart technique). In this paper we draw graphs for various states of India as well as for the country Bolivia to get a clear idea about the IPV.

KEYWORDS: IPV, Statistical Quality Control, Bolivia, Violence, Categories

INTRODUCTION:

In the most recent times human life becomes more advanced, civilized from social and economic side. But apart from this luminous side we have a darken side too in human's social life. And that is the many types of violence occur in human's life. One of the most common forms of violence against men/women is performed by their intimate partners. This type of violence is called Intimate Partner Violence (IPV). IPV is a very common but burning issue in the world including India. Though it is a burning issue but it remains very much under reported. For this reason it is tough to get rid of this burning problem of the society. IPV mainly are of three types, (i) Emotional Violence, (ii) Physical Violence, (iii) Sexual Violence.

Emotional Violence: In this type of violence intimate partner mainly abuse their associates in front of others so that they get embarrassed and desire to have total control on their life. The associates don't have their own likings or disliking. They always humiliate their associates in front of others. They control their associates by economically as well as socially. And in many cases partners are separated them from their own family which is very painful for the associates. Though the associates have their own income but they don't have the permission to spend it by themselves. As a result they don't spend their life freely.

· Physical Violence: This is one of the most cruel violence that is done by an intimate partner on their associates. In this kind of violence intimate partner physically hurt their associates. They harm their associates by beating, slapping or pushing. Sometimes partners use weapon for injuring the associates. For this kind of violence associates face many physical problems even for the life time. They cannot get back their normal life.

· Sexual Violence: This violence is not only injured the associates physically but also mentally. In this kind of violence intimate partners may apply physical force on their associate to engage in a sexual act against their will. These people face many sexual problems. For this type of violence women suffer most. Unwanted sexual life may led to HIV, many sex oriented diseases. Also they face premature pregnancy which dangerously effect on their health. For this kind of violence women sometimes break down physically as well as mentally.

In IPV both men and women suffer from horrendous physical and mental pain but most of the times women suffer severely. Though now a day’s women protest against this but the number of this kind of protest is very few. It is very much under reported as it is a violence which is caused by intimate partners so it is tough to explain in front of people. IPV has not only effect on men or women's life but it has a bad effect on their children too. They suffer from a painful mental stress without any reason. This type of violence has a bad impression on their childhood and has a long term effect on their whole life.

Uthman Olalekan A et al. (2009) worked on IPV in Sub Sharahan African countries. They found out the factors associated with attitudes towards IPV against women. Because they thought the rate of IPV is very high there. They used Pear-son's chi-squared test for analyzing contingency tables and performed random-effects model to incorporate between-country heterogeneity. They used multiple logistic regression models for finding the attitudes of men and women towards Intimate Partner Violence against womens. Country heterogeneity was assessed by using the Cochran Q test and the I2 statistic. Gage. Anastasia.J. et al. (2011) found the child marriage is one of the most influential cause for IPV. They found that whether any observed associations present between child marriage and IPV through regression analysis.Tumwesigye Nazarius Mbona et al. (2012) worked on a particular aspect of IPV. They worked on Physical Intimate Partner Violence (PIPV). They said that alcohol drinking problem among sexual partners was the main reason for Physical Intimate Partner Violence (PIPV). It is known from women's reports that their partner got drunk sometimes or often and served as the main factor of PIPV. For the analysis purpose they used regression analysis. Salihu HM et al. (2012) notice that one of the most cruel way of violence on female is “Female Genital Mutilation"”(FGM). For this purpose researcher used wald chi-square tests of independence to compare differences in socio demographic characteristics between the FGM and non-FGM groups. Palermo Tia et al. (2013) worked on another cruel part of violence i.e. Gender Based Violence (GBV-“Gender Based Violence” and “Violence against women” means the same thing. That is violence occurs on women and girls by men).This is performed by an intimate partner. Actually gender-based violence (GBV) is widespread globally and has myriad adverse health effects but is vastly under reported. They performed regression analysis and multivariate logistic regression for their research purpose. Gracia Enrique et al. (2015) analyzed the influences of neighbourhood-level characteristics on small-area variations in IPV risk by using Bayesian spatial random-effects modelling.

In the above mentioned papers, it has been found that researchers work on IPV from various corners. In some papers, researchers work over the cause and effect analysis of IPV and search for the reasons for occurring of IPV and effects on victim's life. Now we need a process for understanding the IPV situation in our country as well as different states in India.

2. Model: In this paper we are going to find out the current situation of IPV in India as well as in a South American country Bolivia. We had found out this situation on the basis of Statistical Quality Control (SQC). SQC means various statistical methods that can be used for maintaining and also monitoring of quality of products. We find out this situation by using control chart. There are many types of control charts in SQC. But for our case “Control chart for number defective” is appropriate when the standard is not given. We can use this method when the quality characteristic is an attribute, and each product is marked as defective or non-defective. Now if we want to know whether the process is in control or not, we have to determine whether the population fraction defective “P” is the same for all sub-groups. The

process may be based either on the number of defectives, say, “d” in the sample or on the fraction defective, in the sample, where n denotes the number of products examined per sub-group. If no information available for \P" then it will be estimated from the samples themselves.

The control lines are

Lower Control Limit, LCL =

Central Line, CL =

Upper Control Line, UCL =

Here, , where i = 1(1)m. Where m is the number of subgroups.

We use SQC since the quality characteristic is an attribute. We assume that the sampled men and women treated in our case as the products. Some questions are asked to the samples. Each unit of the sample has two options to record of their answer either “No” or “Yes”. Here the option “No” is treated as the number of defectives in each sample. Here we can find out the situation of IPV by checking the number of outliers. For this reason we use Control chart for number of defectives.

To form this type of chart we first have to calculate the Lower Control Limit (LCL), Central Line(CL), Upper Control Limit(UCL). In this case outliers means the data that are falling below the LCL.

For our case

P= Population fraction = The option”No”.

d= Number of times option”No” occur out of the total questions.

n= Number of questions.

m= Number of samples i.e. here it is the total number of sample men and women.

p_i = d_i=n= Fraction defective in the ith sample.

Let, X= A random variable denote the number of times “No” occur out of n trials. As we mentioned before that we use “control chart for number defective”. And in our context we consider the option “NO” as the number of defectives. We mentioned that here “No” option indicate that there is no case of IPV. That means as much data fall between CL and UCL, then we can say IPV situation is better on those cases. As UCL is the upper control limit of “Number of defective” (in our case option “No”). This picture tells us that a major amount of people choose option “No” i.e. there is no violence. Likewise if maximum data falls between CL and LCL then the situation of IPV is bad. And if maximum data fall below the LCL then the IPV situation is worst there. This indicate that a few amount of people choose option “No” i.e. there is a maximum amount of people who choose option “Yes” i.e. IPV occurs with them.

Then, X follows binomial distribution with parameters n and p.

3. Data Analysis: For our study we use “Women data set” from NFHS-3 (National Family Health Survey-3). NFHS is refer to as the “Demographic Health Survey” or DHS in other national context and is conducted regularly in many countries to obtain population based estimates of major health threats. After reviewing earlier published papers on this topic we found there are many types of questions for identifying IPV. From our data set we take the following 8 questions. We choose these particular questions as they covered all type of questions from reviewed papers. NFHS-3 is a big data set. It almost contains 124385 data. We take 69388 data (for India) for our work. We work on this data because here we take 8 questions from the data set, which covered all three types of violence as mentioned before and these questions have not any missing values. Besides for whole India we also work on the states separately. So that we can compare the IPV situation among different states in India separately and with whole India also.

3.1. India: Here we have 8 questions. These 8 questions treated as IPV. Each question has 5 categories. The questions and categories are as follows

a) Spouse ever threatened her with harm.

b) Spouse ever pushed shook or throw something.

c) Spouse ever slapped.

d) Spouse ever punched with fist or something harm full.

e) Spouse ever kicked or dragged.

f) Spouse ever tried to strangle or burn.

g) Spouse ever threatened or attacked with knife or other weapon.

h) Spouse ever twisted her arm or pull her hair.

and categories are

0 = No.

1 = Often during last 12 months.

2 = Some times during last 12 months.

3 = Not in last 12 months.

4 = Yes but currently a widow.

Here we have 5 categories for each question but we divide these categories into 2 categories since we consider it is a binomial case. That means we arrange these categories as

0 = “No”.

1 = 2 = 3 = 4 = “Yes”.

4. Quantification of number of defectives: Here we take these 8 questions as IPV quantifier. We mentioned before that we divide every question in two 2 categories “No” and “Yes” Now if anybody choose option “0” then the people does not face IPV but if he/she choose any one option from 1,2,3 and 4 then we can say they suffer from IPV.

5. Analysis result: The various result that we found is as follows

Table 1: Results on different states

States (1)	Total no of Data (2)	LCL (3)	CL (4)	UCL (5)	Data between CL and UCL (6)	% (6) (7)	Data on and below LCL (8)	% of (8) (9)
Manipur	2170	4.58	7.17	9.76	1249	57.56%	125	5.76%
Bihar	2086	3.06	6.43	9.80	1309	62.75%	220	10.54%
Assam	2260	4.41	7.09	9.78	1424	63.00%	222	9.82%
Orissa	2579	4.27	7.03	9.80	1705	66.11%	274	10.62%
Andhra Pradesh	4276	4.46	7.12	9.77	2871	67.14%	425	9.94%
Chhatisgarh	2093	4.45	7.11	9.78	1443	68.94%	215	10.27%
West Bengal	4026	4.53	7.15	9.76	2788	69.24%	414	10.28%
Tamilnadu	3836	3.76	6.79	9.83	2694	70.23%	309	8.05%
Harayna	1678	4.70	7.22	9.74	1134	71.86%	148	9.37%
Uttranchal	1604	4.84	7.27	9.71	1158	72.19%	117	7.29%
Maharasthra	5127	5.07	7.36	9.66	3703	72.22%	462	9.01%
Gujrat	2227	5.00	7.34	9.67	1613	72.43%	219	9.83%
MP	3802	4.00	6.91	9.82	2821	74.20%	415	10.91%
Punjab	1918	4.90	7.30	9.70	1428	74.45%	157	8.18%
Jharkhand	1720	4.16	6.98	9.81	1284	74.65%	187	10.87%
Rajasthan	2242	4.06	6.94	9.82	1678	74.84%	243	10.84%
Tripura	1099	3.87	6.85	9.83	825	75.07%	110	10.00%
UP	6504	3.98	6.90	9.82	4901	75.35%	510	7.84%
Arunachal Pradesh	943	4.13	6.97	9.81	720	76.35%	97	10.28%
Mizorum	937	5.53	7.53	9.53	717	76.52%	56	5.98%
Karnataka	3450	5.16	7.4	9.64	2738	79.36%	296	8.58%
Delhi	1891	5.62	7.56	9.50	1529	80.86%	120	6.34%
Goa	1691	5.57	7.54	9.51	1398	82.67%	119	7.04%
Kerala	1983	5.70	7.58	9.47	1653	83.36%	115	5.80%
Nagaland	2041	5.95	7.66	9.37	1707	83.63%	94	4.60%
Sikkim	1119	5.64	7.56	9.49	956	85.43%	82	7.33%
Meghalya	1039	6.01	7.68	9.34	900	86.62%	82	7.82%
Jammu and Kashmir	1443	5.89	7.64	9.39	1266	87.73%	88	6.09%
Himachal Pradesh	1704	6.55	7.82	9.08	1600	93.89%	64	3.75%

Here LCL = Lower Control Limit, CL = Central Line, UCL = Upper Control Limit.

Graphs: For the above results the corresponding graphs are given below.

Bolivia: We are also interested to look for the IPV situation in another country. And for that reason we take \Bolivia" which is a South American country. In this case we use a couple data set (2008). We took this data from DHS site. After eliminating all the missing values we work with 2714 data. In this case we have 5 questions which are the quantifier of IPV. The questions are as follows, 1.Partner pushed or pinched respondent.

2. Partner beat or kick respondent.

3. Partner beat her with an object.

4. Partner tried to strangle or burn her.

5. Partner tried to force sex with her.

Like before here also every question has 5 categories but we divide them in two categories. The categories are as follows

1. ”No”, 2.”Yes often”, 3.”Yes a few times”, 4.”Yes one times”, 5.”Don’t know”.

Now we organize them in two like

1 = “No”.

2=3=4=5 = “Yes”.

The result which we found from Bolivia is as follows

Table 2: Results on different countries

States

(1)

Total no of Data

(2)

LCL (3)

CL (4)

UCL (5)

Data between CL and UCL

(6)

% (6)

(7)

Data on and below LCL

(8)

% of (8)

(9)

India

69388

4.61

7.18

9.75

47425

68.35%

5994

8.64%

Bolivia

2714

1.36

4.02

6.68

1629

60.02%

269

9.91%

The corresponding graph of the country Bolivia is given below

7. CONCLUSION:

It is noticed that there are the states viz. Manipur, Bihar, Assam, Orissa, Andhra Pradesh, Chhattisgarh, and West Bengal are relatively less a affected in respect of IPV, whereas the states Tamilnadu, Haryana, Uttaranchal, Maharashtra, Gujarat, MP, Punjab, Jharkhand, Rajasthan, Tripura, UP, Arunachal Pradesh, Mizoram, and Karnataka are better over the above mentioned states as the index for IPV lies between 70.23% to 79.36% within CL and UCL (Table 1). And the states viz. Delhi, Goa, Kerala, Nagaland, Sikkim, Meghalaya, Jammu and Kashmir and Himachal Pradesh are relatively in well acceptable position as the index lies between 80.86% to 93.89% within CL and UCL and IPV is assumed to be negligible. Over all figures for India is 68.35% within CL and UCL indicating comparatively a low intimate partner violence prevailing in these areas.

Among all the states in India we can observe that the IPV situation is in a bad condition in Manipur as only 57.56% data lies between CL and UCL compare to other states. Where Himachal Pradesh is in a good situation as 93.89% data lies between CL and UCL which is very good. Now from the graph 17 we can see that the difference between the two graphs is clear. In the graph of Manipur we can see that the data between CL and UCL is not the maximum amount. Where as in the case of Himachal Pradesh a large amount of data lies between CL and UCL. Since we have lots of data for both the states, so we cannot see those data distinctly in the graph. They are clustered.

We have similar data for Bolivia. Though situation of Bolivia and India is not exactly same in respect of socio-economic and cultural aspects, we have tried to com-pare the overall IPV situation of two countries. From table 2, we can see that in the case of IPV situation India (with 68.35%) slightly in a better position than Bolivia as IPV index (with 60.02%) lies between CL and UCL.

We have done this work for India (including all the states) and Bolivia. This can be done for any country. And after getting the result we can compare among the different countries to know the IPV situation in all over the world. From that result we can have a clear picture that which country is in a good position and which country need some progress in this case to overcome the problem.

8. REFERENCES:

Decker Michele R, Seage George R, et al. (2009): Intimate Partner Violence Functions As Both A Risk Marker And Risk Factor For Women's HIV Infection: Findings From Indian Husband-Wife Dyads, J Acquir Immune De c Syndr. 2009 Aug 15; 51(5): pp- 593600.

Dillon Gina, Hussain Rafat, et al(2012): Mental And Physical Health And Intimate Partner Violence Against Women: A Review Of The Literature ,In-ternational Journal of Family Medicine,Volume 2013, Article ID 313909, 15 pages.

Gage J. Anastasia and Hotchkiss David, Godha Deepali (2012): Association between Child Marriage and Reproductive Health Outcomes: A Multi-Country Study of Sub-Saharan Africa, presented at the Annual Meeting of the Population Association of America, 2-5 May, 2012, San Francisco, California, 1-28 pages.

Gonza Lez D Gil, Cases C Vives, et al. (2006).:Alcohol And Intimate Partner Violence: Do We Have Enough Information To Act, European Journal of Public Health, Vol. 16, No. 3,pp-278-284.

Gracia Enrique, Antonio Lpez-Qulez, et al (2015) : The Spatial Epidemiology of Intimate Partner Violence: Do Neighborhoods Matter? , American Journal of Epidemiology, DOI: 10.1093/aje/kwv016,pp-58-66.

Hossain Mazeda, Zimmerman Cathy, et al. (2014): Working With Men To Prevent Intimate Partner Violence In A Con ict-A ected Setting: A Pilot Cluster Randomized Controlled Trial In Rural Cte d'Ivoire, BMC Public Health BMC series,DOI: 10.1186/1471-2458-14-339.

Joyner Kate, Mash Robert James.(2011): The Value Of Intervening For In-timate Partner Violence In South African Primary Care: Project Evaluation, BMJ Open, 1(2), doi:10.1136/bmjopen-2011-000254.

Koen Nastassja and Wyatt Gail E, et al. (2014): Intimate Partner Violence: Associations With Low Infant Birth Weight In A South African Birth Cohort, Metab Brain Dis, 29(2), pp- 281299.

Lee Minjee, Stefani Katherine M and park Eun-Cheol(2014):Gender-Specic Differences In Risk For Intimate Partner Violence In South Korea, BMC Public Health BMC series 14:415,DOI: 10.1186/1471-2458-14-415.

Meeker Dominique, Pallin Sarah C and Hutchinson Paul (2013):Intimate Partner Violence And Mental Health In Bolivia, 13:28

Palermo Tia, Bleck Jennifer and Peterman Amber (2013): Tip Of The Ice-berg: Reporting And Gender-Based Violence In Developing Countries, Amer-ican Journal of Epidemiology , Vol- 179 (5), pp- 602-612

Salihu HM, August EM, et al. (2012):The Association Between Female Genital Mutilation and Intimate Partner Violence, BJOG:An International Journal of Obstetrics and Gynaecology; 119(13), pp- 1597-1605.

Tumwesigye Nazarius Mbona, Kyomuhendo Grace Bantebya, et al (2012): Problem Drinking And Physical Intimate Partner Violence Against Women: Evidence From A National Survey In Uganda, BMC Public Health BMC se-ries12:399, doi:10.1186/1471-2458-12-399.

Uthman Olalekan A, Lawoko1 Stephen and Moradi Tahereh (2009): Factors As-sociated With Attitudes Towards Intimate Partner Violence Against Women: A Comparative Analysis Of 17 Sub-Saharan Countries, BMC International Health and Human Rights BMC series, DOI: 10.1186/1472-698X-9-14.

Received on 08.06.2017 Modified on 15.06.2017

Int. J. Rev. and Res. Social Sci. 2017; 5(2): 93-102 .

DOI: 10.5958/2454-2687.2017.00010.7