Stigma Index

Analyzing the data

Before all this can happen, data capturers (two or more people) need to enter the raw data or original responses and narratives from each of the questionnaires into a computer-based programme called EPI-Info, so that it can be analyzed. The People Living with HIV Stigma Index has been designed to be used with EPI-Info.

When planning the actual data entry process, it is important to bear in mind the following steps in order to get the raw data collated:

• Install the statistical programme on a computer

EPI-Info is a free software package that contains statistical facilities for developing a questionnaire; customizing the data entry process; entering the data and; analyzing the data. EPI-Info can be downloaded from the Centre for Disease Control and Prevention (CDC) website .

There are a number of versions available, but the latest version is EPI-Info™ Version 3.4.3. EPI-Info is designed for users who only have basic computer skills and gives step-by-step instructions to complete the basic tasks.If you are using EPI-Info for the first time, it is strongly advised that you download the user manual .

• Download the database for the questionnaire in EPI-Info

A centralized database has been created in EPI-Info. Each country team will be using a similar database and the information will also be stored in a central database hosted by IPPF.

• Enter the data

When planning the data entry process, consideration should be given to the amount of time needed to complete this task. From our experience, it takes approximately 20 minutes to enter the data from each questionnaire.

• Analyze the data

This component allows you to access the data from your data tables to perform the analysis. In terms of analysis, EPI-Info will be able to produce lists, frequencies, cross tabulations and many other statistical tasks that will assist you to extract useful findings.

The analysis process can help you:

• describe the experiences of the whole data set;
• make comparisons between groups;  
• verify data.

For example, Section 1 of the questionnaire describes the interviewees in terms of their age, gender, education, employment status and household income. When this information is compared to the responses in Sections 2 and 3 (which focus more on interviewees’ experiences of stigma and discrimination, and their experiences of HIV testing, disclosure and access to services), interesting associations or connections might emerge. These could be, for example, between the educational level and income of interviewees, the nature of a relationship they are in and their access to services. A programme like EPI-Info can assist in exploring associations like these and helps manage a big data set.

 It can also help the research team compare answers between different questions to help verify (or confirm) that the data provided by interviewees are consistent or reliable across themes in the questionnaire. As an example, in Section 1:

• Question 9 (the interviewee’s employment status)
• Question 12 (whether they live in an urban or rural area)
• Question 13 (what the average income of the household is per month) and
• Question 14 (whether the household ever runs out of money to buy food or not)

These can be looked at as a group of questions. When one puts all the responses of one interviewee together, we would expect to be able to build a picture of the life or lifestyle of that person. For example, someone who is in fulltime employment will (most likely) live in an urban area (as there is improved access to employment) and have a relatively comfortable household income. This person is likely to never run out of money for food. Another interviewee may be doing part-time or casual work and really struggle to consistently provide an income for their household and as such be more vulnerable to not having enough money to pay for basic resources like food.