Evaluation of data screening methods in surveys of adolescents' drug use.

Farrell, Albert D.; Danish, Steven J.; Howard, Catherine

doi:10.1037/1040-3590.3.2.295

Cited by 35 publications

(16 citation statements)

References 10 publications

(8 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is notable that these studies have found very low self-reported rates of use of these fictitious substances. Petzel et al [ 206 ], for example, found that 4% of his sample of high school students reported the use of the nonexisting drug “bindro.” They also found that those who reported the use of a nonexistent drug also reported more use of all other drugs included in their survey, compared to those who indicated, correctly, that they did not use “bindro.” Others have reported similar findings when asking survey respondents about the use of nonexistent substances [ 202 , 207 – 209 ]. Of course, it may be that heavy drug users just assume, incorrectly, that they have used all available substances at one time or another in their past.…”

Section: Errors Of Measurementmentioning

confidence: 53%

“…Investigators also use a variety of techniques to screen completed substance use questionnaires for inclusion in final data files. Farrell and colleagues [ 207 ] examined the effects of excluding respondents (1) who provided a large number of inconsistent answers and (2) who reported use of a fictitious substance. The effects of excluding these responses on prevalence estimates were considered to be minimal, although they cautioned that exclusionary criteria should be used carefully in order to avoid producing nonrepresentative results.…”

Section: Errors Of Measurementmentioning

confidence: 99%

See 1 more Smart Citation

Sources of Error in Substance Use Prevalence Surveys

Johnson

2014

International Scholarly Research Notices

151

118

View full text Add to dashboard Cite

Population-based estimates of substance use patterns have been regularly reported now for several decades. Concerns with the quality of the survey methodologies employed to produce those estimates date back almost as far. Those concerns have led to a considerable body of research specifically focused on understanding the nature and consequences of survey-based errors in substance use epidemiology. This paper reviews and summarizes that empirical research by organizing it within a total survey error model framework that considers multiple types of representation and measurement errors. Gaps in our knowledge of error sources in substance use surveys and areas needing future research are also identified.

show abstract

Section: Errors Of Measurementmentioning

confidence: 53%

Section: Errors Of Measurementmentioning

confidence: 99%

Sources of Error in Substance Use Prevalence Surveys

Johnson

2014

International Scholarly Research Notices

151

118

View full text Add to dashboard Cite

show abstract

“…In February 1998, the Florida Department of Health undertook a self-administered classroom survey of nearly 23,000 public middle (grades [6][7][8] and high (grades 9-12) school students. The study was conducted in cooperation with the Florida Department of Education, with assistance from the Office on Smoking and Health and the Division of Adolescent and School Health at the Centers for Disease Control and Prevention in Atlanta, Georgia.…”

Section: Methodsmentioning

confidence: 99%

Editing Data: What Difference Do Consistency Checks Make?

Bauer

Johnson²

2000

American Journal of Epidemiology

View full text Add to dashboard Cite

In 1998, the Florida Department of Health undertook a self-administered school-based survey of tobacco use, attitudes, and behaviors among nearly 23,000 public school students in grades 6-12. The survey design did not use skip patterns; therefore, students had multiple opportunities to contradict themselves. By using examples from the high school portion (grades 9-12) of the survey, the authors examined five possible approaches to handling data inconsistencies and the effect that each has on point estimates. Use of these approaches resulted in point estimates of current cigarette use ranging from 25.6% to 29.7%. The number of missing respondents varied from 33 (less than 1%) to 1,374 (13%), depending on which approach was used. After stratification by gender and race, the prevalence estimates changed marginally for girls but strikingly for boys. Non-Hispanic White students were substantially more likely than non-Hispanic Black students to report current cigarette use, but the magnitude of this difference varied significantly according to the analytical approach used. The approach used to check data consistency may influence point estimates and comparability with other studies. Therefore, this issue should be addressed when findings are reported.

show abstract

“…3 With few exceptions (e.g., Farrell, Danish, & Howard, 1999), the ISD or similar response variance indexes have not yet been used to identify random responding. They have, however, been used in related psychometric fields that assess itemresponse accuracy (Charter, 2000), personality stability (Asendorpf, 1992), and test reliability (Sturman, Cribbie, & Flett, 2009).…”

Section: Participantsmentioning

confidence: 99%

The inter-item standard deviation (ISD): An index that discriminates between conscientious and random responders

Marjanovic

Holden

Struthers

et al. 2015

Personality and Individual Differences

102

View full text Add to dashboard Cite

Although random responding is prevalent and increases Type II errors, most psychologists avoid trying to identify it because the means to do so are extremely limited. We propose the inter-item standard deviation (ISD), a statistical index of response variance, is suited for this task. We hypothesized that random responders produce large ISDs because they respond to items all over a measure's response range, whereas conscientious responders produce small ISDs because they respond to items more consistently. We administered a questionnaire containing the NEO-FFI-3 and an embedded validity scale to 134 university students. Another 134 responders were created using a random number generator. For all 268 responders, the ISD was calculated for each of the NEO-FFI-3 0 s five subscales and an aggregated ISD was calculated by averaging the five ISD indexes. Results showed that (1) random responders produce significantly larger ISDs than conscientious responders, (2) the ISDs were strongly correlated with the embedded validity scale and with one another, and (3) the ISDs correctly identified responders with greater than 80% classification accuracy. The mean ISD yielded greater than 95% classification accuracy. This study shows that responders can be identified by quantifying inter-item response variance.

show abstract

Evaluation of data screening methods in surveys of adolescents' drug use.

Cited by 35 publications

References 10 publications

Sources of Error in Substance Use Prevalence Surveys

Sources of Error in Substance Use Prevalence Surveys

Editing Data: What Difference Do Consistency Checks Make?

The inter-item standard deviation (ISD): An index that discriminates between conscientious and random responders

Contact Info

Product

Resources

About