Descriptive analysis (Logistic Regression)

Descriptive analysis (Logistic Regression)

Descriptive analysis (Logistic Regression)

Introduction

An approach to logistic regression, as one of the most used multivariate statistical techniques in the last decades has passed, was made to recommend correct use of STI. Practical questions as the number of subjects necessary for its application, the situations in which it should be used, the type of variables to which it may be applied, the way it may be included in the model, the interpretation of the results, etc., were taken into consideration. Practical questions as the number of subjects for its application, the situations in which it should be used, the type of variables to which it may be applied, the way it may be included in the model, the interpretation of the results, etc., were taken into consideration. It is concluded that the application of logistic regression is very useful in any field of psychological research; we determine the effect of a group of age groups, sex and education variables, potentially considered as influential, on the occurrence of the process.

A survey has been conducted to find out the impact of Education, Marital status, Age and Gender on the Physical violence and Emotional Abuse.

Independent variables

Education (EDUPR5R) has been subdivided in (No HS, HS, Diploma and University)

Marital Status (MARSTATR) (Married, Widowed, Divorced/separated and single)

Age (AGEGR10) (18-24, 25-34, 35-44, 45-54, 55-64 and 65+)

Gender (SEXR) (Men and Women)

Dependent Variables

Physical violence (During the past 5 years,: ... did he/she slap you PSX_Q140R) (During the past 5 years,: ... did he/she kick you, bite you, or hit you with his/her fist PSX_Q150R) (During the past 5 years,: ... did he/she beat you PSX_Q170R) (During the past 5 years,: ... did he/she choke you  PSX_Q180R)

Emotional abuse (He/She put you down or called you names to make you feel bad EFX_Q220R)

(He/She was jealous and didn't want you to talk to other men/women EFX_Q230R)

(He/She harmed, or threatened to harm, someone close to you EFX_Q240R)

(He/She demanded to know who you were with and where you were at all times EFX_Q250R)

There is no doubt that logistic regression is a statistical tools to better capacity for data analysis in clinical research and epidemiology, hence it can be very widely use. The primary objective solve by this technique is to model how it affects the probability of occurrence of an event, usually dichotomous, the presence or absence of various factors and the value or the same level. It can also be used to estimate the probability of occurrence of each of the possibilities of an event with more than two categories (poly-tomous). Everyone knows that this type of situation is addressed through regression techniques. However, the methodology of linear regression is not applicable because now the response variable has only two values ??(we focus on the dichotomous case), such as YES/NO. In this study we are analysing the effects of different variables on physical violence and emotional abuse.

Sometimes it is thought that the influence of one variable on the probability of occurrence of the event is modified depending on ...