different terms have been used to describe this phenomenon, including poor response, nonresponse, and negative response to exercise interventions. The most common definition used is a lack of change of an outcome of interest in the expected direction ( Bouchard, 1995 ; Bouchard & Rankinen, 2001 ; Bouchard
Mary O. Whipple, Erica N. Schorr, Kristine M.C. Talley, Ruth Lindquist, Ulf G. Bronas and Diane Treat-Jacobson
Jeremy S. Jordan, Matthew Walker, Aubrey Kent and Yuhei Inoue
The failure to adequately address nonresponse issues in survey research may lead to nonresponse bias in overall survey estimates, which can severely restrict researchers’ ability to make inferences to a target population. This study was designed to assess the frequency of nonresponse analyses in articles published in the Journal of Sport Management (JSM). All articles from the years 1987 through 2008 published in JSM (N = 371) were content analyzed based on a previously established coding scheme as well as additional indicators. The results revealed that only a small number of articles reported the use of nonresponse analyses as a means to control for nonresponse error.
Nicola W. Burton, Gavin Turrell and Brian Oldenburg
This study assessed item nonresponse (INR) in a population-based mail survey of physical activity (PA).
A questionnaire was mailed to a random sample, with a 57% response rate (n = 2532). The magnitude and type of PA INR and the association with sociodemographic variables was examined using logistic regression.
Among survey respondents, 28% had incomplete PA data; 11% missed 1 item, 11% missed 2 items, and 5% missed all 3 items. Respondents missing 3 items tended to be female, less educated, low income, in poor health, and current smokers. The walking item was missed by 8% of respondents, and 18% and 23% missed the vigorous-intensity and moderate-intensity PA items respectively. These groups were sociodemograpically different from those without INR. Incomplete PA data was also associated with sociodemographic INR.
Mail surveys may underrepresent individuals insufficiently active for health, in particular those of low socioeconomic position.
Paula C. Fletcher and John P. Hirdes
This paper examines factors associated with physical activity and health status among the 796 subjects aged 55 and older who appear in both the 1981 Canada Fitness Survey (CFS) and The Campbell’s Survey on Well-Being (CSWB), a longitudinal follow-up to the CFS. The CSWB can provide information about changes in physical activity patterns and health between 1981 and 1988. Although nonresponse to the overall survey was low, item nonresponse was problematic in some cases. Approximately 50% of the sample were not assessed on physical fitness measures (e.g., body mass index), while 14% and 38% refused to answer questions concerning alcohol consumption and family income, respectively. Of specific interest are the relationships of physical activity levels and self-rated health with socio-economic status, age, gender, smoking history, alcohol consumption, and measures of body composition.
David L. Porretta, Francis M. Kozub and Fabio L. Lisboa
Articles related to adapted physical activity appearing in professional journals (1984-1998) were analyzed. Of the 111 articles reviewed, 30 (27%), 39 (35%), and 42 (38%) were published during the 1984-1988, 1989-1993, and 1994-1998 time periods, respectively. Two thirds of the studies concerned conditions/demographics/practices rather than attitudes. Only 34 (31%) surveys were mailed as opposed to other forms of delivery (e.g., face to face interviews, telephone, etc.). While validity and reliability reporting increased over the three time periods, in total, only 59 (53%) reported validity and 62 (56%) reported reliability. A sample frame was clearly identified in only 43 (39%) studies. Only 7 (6%) articles addressed nonresponse bias, a critical element in survey research design. Future investigators need to report validity and reliability, clearly define sample frames, and account for nonresponse bias.
Cora Lynn Craig, Lise Gauvin, Sue Cragg, Peter T. Katzmarzyk, Thomas Stephens, Storm J. Russell, Lloyd Bentz and Louise Potvin
The health benefits of physical activity are substantial; however, the lifetime and environmental determinants of sedentary living are poorly understood. The purpose of this article is to outline the conceptual background and methods of the Physical Activity Longitudinal Study (PALS), a follow-up study of a population- and place-based cohort. A secondary purpose is to report on the success of follow-up procedures.
A rationale for conducting a 20-y follow-up of a nationally representative population- and place-based cohort is developed based on the extant literature dealing with socio-environmental determinants of health and on current advancements in thinking about the determinants of involvement in physical activity. Then, methods of the 2002-04 PALS (n = 2511, nonresponse = 29.8%) that began with the 1981 Canada Fitness Survey are described. Descriptive data pertaining to the success of follow-up procedures are outlined.
There is general consensus around the relevance of examining lifetime and environmental determinants of physical activity involvement. Longitudinal data represent one source of information for disentangling the relative importance of these determinants. Examination of PALS follow-up data show that there was no selection bias for key individual- (physical activity, other lifestyle, health) and area-level (median income, housing) variables, although fewer respondents than nonrespondents smoked or were underweight at baseline. Some demographic groups were under- or over-represented among the eligible cohort, but not among participants.
The social epidemiological perspective emerging from PALS should help policymakers and public health practitioners make strides in changing socio-environmental factors to curb sedentary lifestyles and promote population health.
Nicole J. Chimera, Monica R. Lininger, Bethany Hudson, Christopher Kendall, Lindsay Plucknette, Timothy Szalkowski and Meghan Warren
/34) 97% (33/34) 97% (33/34) 35 1 removed (asked to be removed from study) 34 2 78% (25/32) 97% (31/32) 97% (31/32) 34 2 removed (season ending injury and nonresponse) 32 3 74% (23/31) 97% (30/31) 97% (30/31) 32 1 removed (nonresponse) 31 4 74% (20/27) 89% (24/27) 93% (25/27) 31 4 removed (nonresponse) 27
Justin A. Haegele, Carrie J. Aigner and Sean Healy
adjusted for nonresponse rates and several demographic factors. Analyses were performed separately for children and adolescents. Descriptive statistics for PA, ST, and sleep prevalence rates for children and adolescents with and without VIs were computed. Chi-square statistics were computed to examine
Meera Sreedhara, Karin Valentine Goins, Christine Frisard, Milagros C. Rosal and Stephenie C. Lemon
status, LHD characteristics, and each of the 8 outcome measures. We conducted 2 sensitivity analyses to compare LHD responders and nonresponders and LHDs with complete data against those with incomplete data using chi-square tests to assess nonresponse and selection bias. Continuous variables were
Matthew Katz, Bob Heere and E. Nicole Melton
, another email request was distributed by the athletic department. The two e-mail requests for participation yielded 138 usable surveys, for a response rate of 18.2%. We assessed nonresponse bias by comparing the responses of the early and late responders. According to Rogelberg and Luong ( 1998 ), late