Thursday, November 10
General
Thu, Nov 10, 10:00 AM - 10:40 AM
Promenade Upper
Thursday Poster Session, Part 1

Survey Quality Predictor (303307)

*Diana Zavala-Rojas, European Social Survey, Universitat Pompeu Fabra 

Keywords: measurement quality, SQP, validity, reliability, measurement error, data quality

This poster presents the program Survey Quality Predictor (SQP) 2.1, an online program to get predictions of the quality of survey questions available for free at sqp.upf.edu.

SQP 2.1 consists on a large database of survey questions with quality estimates and predictions. Quality estimates are obtained from Multitrait-Multimethod (MTMM) analyses, while quality predictions are obtained from SQP 2.1. This large database includes all survey questions from the European Social Survey (ESS) Rounds 1 to 6 and survey questions from a large variety of research fields, in many different countries and about many different topics.

Using this program, the users can obtain a prediction of the quality of new or currently available survey questions including reliability, validity and quality coefficients with confidence intervals and standard errors, and suggestions for improving them, in many different languages and for more than 20 countries. The only effort needed is to introduce the survey question and code its formal characteristics. The coding process in SQP 2.1 consists of 30 to 60 formal characteristics of the survey question, depending on its complexity. Examples of such characteristics are: the domain, the concept, the social desirability, the number of points in the answer scale, the presence of instructions for respondents or interviewers, etc.

Thus, SQP 2.1 is a very powerful tool both at the stage of questionnaire design, before data collection, in order to improve the survey questions forms, and at the stage of data analysis, after data collection, in order to correct for measurement errors.

The poster explains what is behind SQP 2.1, a meta-analysis of thousands of reliability and validity estimates obtained through MTMM experiments. Furthermore, it explains what you can do using this program, and how to proceed to do it.