Survey Measurement & Analysis

Semester Spring
Level 11
Credit Value 20
Module Co-ordinator Dr Simon McCabe
Contact Hours 11 x 2 hour Lectures and 9 x 2 hour seminars

40% of course grade: Exam and 60%: Policy/business intervention proposal.  The report will involve the use of the European Social Survey latest round data to formulate a hypothesis or set of hypotheses on the statistical determinants of a key outcome of relevance to behavioural science (e.g. well-being, trust, job-satisfaction).  Students will be required to specify a regression module to test these hypotheses, report the findings, draw conclusions, and outline the limitations of the research and potential future studies.


The module aims to:

  • Provide an understanding of survey design and the principles for sound construction of self-report survey measures.
  • Develop an awareness of limitations in the use of subjective measures as dependent variables in standard regression designs (e.g. intercultural incomparability, use of non-probability samples) and the methods that have been used to overcome these (e.g. anchoring vignettes, high-frequency measurement).
  • To build an understanding of the self-reported and subjective measures that can be incorporated in economic and policy studies as explanatory variables explaining outcomes such as health and education.
  • To provide an understanding of novel measurement tools that have been produced chiefly in psychology (e.g. real-time activity tracking, Day Reconstruction Method) but are increasingly being incorporated into economics.
  • Expose students to the use of secondary datasets to study constructs and research questions which exist at the interface of economics and psychology. Such secondary datasets have started to contain a large number of constructs (such as personality at multiple time points) giving much scope for researchers in behavioural science to benefit from these resources.
  • Provide a background in biosocial surveys. Increasingly detailed assessments of biological markers of human functioning are now an important component of large-scale government surveys in the social sciences (e.g. NCDS, Add Health). This component of the module will address the structure of the primary biological measures examined in social surveys and the potential biological basis of economic decisions. 
  • Provide training in the basic operations of Stata, including how to work with Do-files, using outreg2, reshaping data.
  • Provide training in OLS regression, Probit regression and computing marginal effects & longitudinal analysis.

 Learning Outcomes

By the end of the module, students are expected to have achieved the following:

  • Become familiar with basic econometric methods for the analysis of survey data.
  • Developed a comprehensive overview of differential item functioning, namely what happens when respondents to survey questions use different criteria for judging what the question means (King et al., 2004) and understand how anchoring vignettes and hierarchical regression modules can take into account these errors.
  • Developed an awareness of methods of measuring behaviour and psychological functioning in daily life and their application to business and policy.
  • A recent literature (e.g. Borghans, Heckman, Duckworth and ter Weel, 2008) has examined how to integrate constructs from psychology into understanding economic outcomes. This literature is rapidly becoming one of the major areas in fields such as health economics and education economics. However, there are many issues with using variables such as personality in econometric functions. We examine new statistical designs for incorporating such measures. 
  • Achieved an understanding of how to utilize techniques for the measurement of experience and activities in daily life and how such measures can be implemented to study key questions in business and policy domains.
  • Have an understanding of the measurement of biological functioning and genetic variation in large scale surveys and how these factors can be used to understand economic behaviour and policy-relevant outcomes.

This module information is representative of what is included in the module in a given year. Details of actual reading, lectures and coursework may vary year to year and will be available at the beginning of the semester.


© University of Stirling FK9 4LA Scotland UK • Telephone +44 1786 473171 • Scottish Charity No SC011159
Portal Logon