Physical exercise and social inequality in Norway – A comparison of OLS and quantile regression analysis

Abstract

This file was dowloaded from the institutional repository Brage NIH -brage.bibsys.no/nih Seippel, Ø. N. (2015). Physical exercise and social inequality in Norway -A comparison of OLS and quantile regression analysis Ørnulf SeippelNorwegian School of Sport Sciences, Oslo, NorwayAbstract: About two thirds of the Norwegian population exercise less than recommended by public authorities. Traditional studies of exercise and social inequality using OLS-regression give us information on the average person (conditioned mean value of a distribution), but might ignore information useful for understanding the situation for those, as in the Norwegian case, less physically active. In this article quantile regression -addressing units at various quantiles, not only the mean -is used to study social inequality related to physical exercise. Three types of exercise are studied -exercising locally, associational sport, fi tness exercise -in light of social background variables. The crux of the analyses is an understanding of how the dependence of various independent variables differs across quantiles and that these results in several cases also differ from what OLS-regression tells us. The data applied are Norwegian (ISSP 2007).

Social integration in sports clubs: individual and organisational factors in a European context

“…These findings mainly confirm results from previous studies (e.g. Seippel, 2005Seippel, , 2015Vandermeerschen, Vos, & Scheerder, 2015;Østerlund, 2014;Østerlund & Seippel, 2013). Less consistent were the results obtained in relation to the integration of minorities.…”

“…Previous research also supports this picture (Baur & Braun, 2003;Hovemann & Wicker, 2009;Seippel, 2015;Van Tuyckom, Scheerder, & Bracke, 2010;Vandermeerschen, Vos, & Scheerder, 2015;Østerlund, 2014;Østerlund & Seippel, 2013). Therefore, we expect that (Hi2) younger people in general are more socially integrated in organised sports than older people.…”

“…One could think that social class (measured as education and/or income) is conducive to integration in sports in cases where the sport has a class character: being costly, demanding time or having a specific class culture. Studies of sports participation have found educational level to correlate positively with levels of physical activity and sports club participation (EU, 2014;Scheerder & Vos, 2011;Seippel, 2015;Studer, Schlesinger, & Engel, 2011) (Andersen & Bakken, 2018). Research conducted among members in sports clubs indicate, however, that once affiliated with a sports club, educational level exerts no or a limited influence on democratic and social participation as well as engagement in community relations.…”

Parental satisfaction with inclusion in physical education

“…QR was initially introduced as a more robust regression analysis with less strict assumptions. However, previous research often used the method mainly because it enables the identification of information about distribution points of the outcome variable other than the conditional mean (Seippel 2015). We used QR to examine whether the explanatory strength of the independent variables varied according to the degree of satisfaction with inclusion in PE on five selected quantiles of distribution, namely, the 0.10, 0.25, 0.50 (median), 0.75, and 0.90 quantiles.…”

A Comparison of Mean-Based and Quantile Regression Methods for Analyzing Self-Report Dietary Intake Data

“…Application of QR to health and behavioral sciences is increasing and could be a valuable statistical tool for health researchers. QR has been used to evaluate the effects of physical activity or dietary intake on varying quantile levels of certain variables, such as BMI [9][10][11][12], waist circumference [13], socioeconomic status [14], and risk factors of disease outcomes including health-related scores and biomarker data [15][16][17][18][19]. A limited number of studies have introduced a QRbased approach specifically applied to behavioral data [20][21][22].…”