Validation Studies Validation

Validation Studies

SHARE THE ARTICLE ON

Table of Contents

Testing Validity in Survey Research

Validity is a very important concept in survey research. It follows the principle that survey questions seek information that in theory can be verified or refuted using an external source. An implication of this principle is that it assumes that validation is appropriately limited to behaviour where an independent record of such behaviour is accessible. This source is generally a set of administrative records that are deemed as the ‘gold standard’ in existing literature. To test validity, researchers will compare self-reported information with this gold standard using some kind of matching process. Based on the results of the match, the self-reports will be classified as either ‘accurate’ or as ‘misreports’. 

Validation studies are used in research to compare the accuracy of a measure with a gold standard measure so as to identify and eliminate bias. Within this article, we will explore the ways in which a measure is validated. 

Download Market Research Toolkit

Get market research trends guide, Online Surveys guide, Agile Market Research Guide & 5 Market research Template

Making the most of your B2B market research in 2021 PDF 3 s 1.png

Validation of Self-Reported Voting Behavior

Validation Studies Validation

A large amount of research has been conducted on the subject of validation of self-reported voting behaviour. These studies are focused on understanding which variables are best suited to validation and on understanding what ‘misreporting’ is. 

Reasons for Misreporting

There are many different reasons respondents may misreport answers, including:

  1. Social Desirability: Respondents may misreport answers so as to present themselves in a more positive light and therefore appear more socially desirable. 
  2. Memory Issues: Respondents with memory issues may misreport the amount or direction of variables unintentionally. 
  3. Source Monitoring: Respondents may be unable to distinguish a specific event from others that are very similar. 

Conditions of the Interview

Voting studies experience an increase in overreporting as the amount of elapsed time from election day increases. This suggests a role for episodic memory. However, research conducted on vote validation studies reflects that the presence of other people does not influence misreporting.

Research on mode effects suggests that survey techniques that have greater privacy are often able to obtain responses on behaviour that are considered sensitive or less desirable. This indicates that self-administered questionnaires are likely to produce higher self-reported levels of behaviours such as smoking than survey methods that involve an interviewer or moderator present. 

See Voxco survey software in action with a Free demo.

Other Methods of Validating Survey Responses

There are many different techniques that can be employed for the validation of survey responses. A common method is to use diaries to record and track events and exhibited behaviours as they occur. Then, the information that is aggregated from samples of diary users can be used to evaluate survey-based measures of the same phenomenon.

On an aggregate level, the rate of bias derived from diaries tends to be low, however, this does increase on an individual level. 

Other Areas of Validation Studies

Validation Studies Validation

There is another form of validation wherein pairs of individuals or groups of individuals are asked about the activities they engage in. Their level of correspondence is then compared in their answers. Such studies have been conducted commonly within the healthcare industry, especially in studies involving partners. 

Research in health care is an important area where the validity of responses must be investigated as the information gathered is used to develop treatment protocols, diagnoses, observation patterns, and medication regimens. 

The Importance of Validation Studies

Validation studies depict the fragility of the data collected through polls and surveys and reflect the number of errors that can exist when studying opinions and behaviours. The levels and types of misreports are influenced by the personal characteristics of respondents and the psychological forces that have an effect on the social and cognitive aspects of the interview process. Characteristics of the interview such as mode of data collection, and the wording and order of questions can also affect self-reports. 

Researchers that study public opinion are interested in understanding the prevalence and size of such effects as they influence the development of models that specify relationships between the variables being investigated.

FAQs on Validation Studies

 Validation studies are used in research to compare the accuracy of a measure with a “gold standard measure” so as to identify and eliminate bias.

Validation studies help researchers identify the errors that exist when studying opinions and behaviours so that these errors can be minimized or eliminated, and research studies can obtain more accurate and valid results.

There are many different reasons respondents may misreport survey responses, including;

  • In an attempt to sound more socially desirable 
  • Due to memory issues
  • They may be unable to distinguish a specific event from others that are similar (source monitoring)

Historically, validation studies have mainly focused on self-reported voting behaviour due to public opinion research being conducted in conjunction with media organizations that are interested in election coverage.

Explore all the survey question types
possible on Voxco

Read more

Hindol Basu 
GM, Voxco Intelligence

Webinar

How to Derive the ROI of a Customer Churn Model

30th November
11:00 AM ET