Using the EquityTool with other data collection platforms

You don’t have to use our web app or DHIS2 to use the EquityTool. The data can be collected using paper forms, CommCare, Magpi, ODK-based tools and most other data collection platforms.

➀ Download the Excel file for your country

The information you will need is in an Excel file. The questions for each country are different, so you will need to download the Excel file for your country. To download the file, go to the country list, click on your country and then click on the ‘Download Excel version’ button. You will need to fill in your details and you will be emailed a zip folder containing your Excel file.

 

➁ Insert the questions from the ‘Questionnaire’ tab to your questionnaire

Inside the Excel file, you will find four tabs. The first is ‘Questionnaire’. These are the questions you should use which can be inserted to your questionnaire. In some countries, the questions are provided in both the local language and English. All the questions and response options should be included and they should not be altered.

Note the ‘Variable name’ column – in the final dataset, the variable corresponding to each question should match the variable name given here. Option 1 should be coded ‘1’, option 2 should be coded ‘2’ and if applicable option 3 should be coded ‘3’.

➂ Collect and enter the data

You can now interview your respondents. Ensure that your sample is adequate – if you would like help with your survey, please get in touch with Metrics for Management about the support services we can offer.

If you collected data on paper forms, these should be entered to a dataset using appropriate data entry software. If possible, use data entry software that has data quality features to help ensure that data entry errors are avoided.

➃ Analyze the dataset

The analysis follows this sequence:

  • Each respondent will get a national wealth index score for each question, depending on which option they chose.
  • The scores for all the questions is added together for each respondent, giving each respondent a total national score.
  • The respondent is then be assigned a national wealth quintile based on their score. Their total score must be greater than or equal to the lower limit of a quintile, and less than the lower limit of the next quintile up.
  • To calculate urban quintiles, the process is the same as for national quintiles using urban scores and urban quintile lower limits.

In the Excel file, you will find STATA commands and SPSS syntax that will implement this analysis automatically, as well as the information you need to conduct this analysis using other software.

If using SPSS

In the Excel file, you will find preprepared SPSS syntax in the second tabs. Ensure you have used the variable names and response option codes as described in step 1. You can then copy and paste the contents of the SPSS syntax tab to the SPSS syntax editor window and run the syntax. If you need urban quintile results, SPSS syntax is included for this purpose.

This should generate your quintile results, showing you the percentage of respondents falling into each quintile:

 

If using STATA

In the Excel file, you will find preprepared STATA commands in the second tabs. Ensure you have used the variable names and response option codes as described in step 1. You can then copy and paste the contents of the STATA do-file tab to a STATA do-file editor window and run the commands. If you need urban quintile results, STATA commands are included for this purpose.

This should generate your quintile results, showing you the percentage of respondents falling into each quintile:

If using other analysis software

If you are not using either of these softwares, you can still analyse the data using the instructions on the ‘Other software’ tab. The scores that correspond to each option are included along with the lower limits of each quintile. Your analysis should follow the following sequence:

  • Create variables that will contain the national scores for each question.
  • Sum these scores to create a total score for each respondent.
  • Assign each respondent to a quintile based on the lower quintile limits provided.
  • Repeat this process using the urban scores and urban quintile lower limits provided for urban respondents.