Surveys construct validity

Research has been carried out to test the psychometric properties of the Safety Attitudes Questionnaire and AHRQ Hospital Safety Culture Survey (Nieva and Sorra 2003 Sexton et al. 2006 Sorra and Nieva 2004). However, there have been important misapplications of survey tools in healthcare. A proliferation of climate surveys now exist, including many where the factor stmcture and construct validity have not been tested. Many hospitals have developed their own bespoke survey tools and these are poorly designed because basic survey design rules have not been followed. One common issue is surveys that do not counter-balance positive and negative statements, thus increasing the risk of response set bias where the... 

Validity refers to whether the smvey instrument measures what it claims to measure. There are three basic types of validity for a measurement scale, each with particular experimental and statistical methodologies for evaluation. These are content validity (do relevant experts agree that the survey appears to measure what it is supposed to measme ), criterion validity (can scores from the survey be used to predict individual behavior or performance ), and construct validity (are the relationships found with the survey consistent with relevant theory and research ). 

The construct validity of a scale is usually evaluated witir tests of convergent and divergent validity. Convergent validity refers to tire extent that other measures of the same construct (for example, a visual estimate of weight) relate to each other. Divergent validity indicates tire extent that scores from surveys unrelated to the construct do not correlate with survey scores related to the construct. In other words, divergent validity implies the extent a particular questionnaire measures special characteristics not measured by other scales. 

The importance of construct validity. It is possible that a direct relationship between a predictor (such as the measure of accident proneness) and a criterion (such as the number of at-risk behaviors or recordable injuries) can be found (predictive validity) without supporting the underlying principle(s) or theory. This would indicate the absence of construct validity. Suppose, for example, an individual could figure out how to answer the survey questions in order to receive a favorable score. Then, construct validity would be questionable, even if criterion validity were high. 

For example, the significant correlation could reflect motivation or intelligence factors rather than actual injury propensity, thus concurrent validity would be shown without construct validity. This kind of hiring survey might select individuals who are most skilled at impression management rather than less injury prone. 

Owens J, Nobile C, McGuinn M, Spirito A (2000) The children s sleep habits questionnaire construction and validation of a sleep survey for school-aged children. Sleep 23 1043-1051... 

The measurement goal in the HIE was to construct the best possible scales for measuring a broad array of functioning and well-being concepts it demonstrated the potential of scales, constructed from self-administered surveys, as reliable and valid tools for assessing changes in health status. It, however, left two questions unanswered Can methods of data collection and scale construction work in sick and elderly populations In addition, could scales that are more efficient be constructed The answer to these questions was the challenge... 

McHomey CA, Ware JE Jr, Raczek AE. The MOS 36-Item Short-Form Health Survey (SF-36) II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care 1993 31 247-263. 

The survey was sent to 446 employees overall in four firms (see Table 21 for distribution between firms). The data collection phase lasted one month for each firm and took place in 2011. After two weeks, a reminder was sent to the employees. 173 responses could be collected, which yielded an overall response rate of 38,8%. Of these responses, 24 were discarded, as they were not filled out to a sufficient degree. This left 149 valid answers for the anafysis. These 149 responses were full responses with no missing data in the focal constructs of the study. Some of the questions, which were sensible (e.g job satisfaction) were optional, and were not provided by all of the respondents. 

Start from the beginning in a process of creating a new facility as the creditability of Stephan s statement is validated. Consider, first, a site survey for ecological considerations, soil testing, and then move into the construction and fitting out of a facility. 

The most prominent survey at the time was the Minnesota Safety Perception Survey. This survey had been developed in the 1980s by a team that included safety professionals from the Association of American Railroads and scientists from the Army s Aberdeen Proving Ground. Dan Petersen served as a consultant to the team, which was led by Chuck Bailey (Bailey and Petersen 1989 Bailey 1997). Bailey s team designed and validated the Minnesota survey, which had subsequently been administered to over 100,000 anployees in a number of companies. Bailey s database included survey results from six chemical companies, giving us a good opportunity for benchmarking. Consequently, we chose that survey. Chuck Bailey made a presentation to our team to explain how the survey had been constructed and how it could be used as a tool for improvement. 

---