PRINCIPLES: Low response rates are common in primary care research. Our study examines the representativeness of respondents in a survey among general practitioners (GPs). One special aim was to evaluate the representativeness of the subgroup of GP teachers for undergraduates (GPTUs) and to investigate the option of a panel of GPTUs. METHODS: The representativeness of the respondents was assessed by the use of pooled public data to compare the respondents and all GPs in the German federal state of Saxony on the basis of socio-demographic and subject-specific characteristics. The representativeness of the GPTUs was examined in the same way. For the analysis, two-sided t-tests and Chi2 tests were used. RESULTS: The total response rate was low (32.87%). The respondents were not a representative sample; in particular, they were more highly qualified than the mean. However, the response rate among the special group of university-associated GP teachers for undergraduates was significantly higher than among other general practitioners. Because of this, the creation of a panel of these GPTUs for further primary care research was investigated. Unfortunately, analysis of this group showed that GPTUs were not a representative sample as they tended to be younger and more highly qualified. CONCLUSIONS: In general it is possible to create a panel of GPTUs to obtain higher response rates, but investigation of the panel's representativeness is definitely required. If the panel is not representative another option is the creation of a stratified sample according to the target population.
PRINCIPLES: Low response rates are common in primary care research. Our study examines the representativeness of respondents in a survey among general practitioners (GPs). One special aim was to evaluate the representativeness of the subgroup of GP teachers for undergraduates (GPTUs) and to investigate the option of a panel of GPTUs. METHODS: The representativeness of the respondents was assessed by the use of pooled public data to compare the respondents and all GPs in the German federal state of Saxony on the basis of socio-demographic and subject-specific characteristics. The representativeness of the GPTUs was examined in the same way. For the analysis, two-sided t-tests and Chi2 tests were used. RESULTS: The total response rate was low (32.87%). The respondents were not a representative sample; in particular, they were more highly qualified than the mean. However, the response rate among the special group of university-associated GP teachers for undergraduates was significantly higher than among other general practitioners. Because of this, the creation of a panel of these GPTUs for further primary care research was investigated. Unfortunately, analysis of this group showed that GPTUs were not a representative sample as they tended to be younger and more highly qualified. CONCLUSIONS: In general it is possible to create a panel of GPTUs to obtain higher response rates, but investigation of the panel's representativeness is definitely required. If the panel is not representative another option is the creation of a stratified sample according to the target population.
Authors: Allanah Li; Shawna Cronin; Yu Qing Bai; Kevin Walker; Mehdi Ammi; William Hogg; Sabrina T Wong; Walter P Wodchis Journal: BMC Fam Pract Date: 2018-05-30 Impact factor: 2.497
Authors: Claudette de Vries; Carine Doggen; Ellen Hilbers; Robert Verheij; Maarten IJzerman; Robert Geertsma; Ron Kusters Journal: BMC Fam Pract Date: 2015-02-05 Impact factor: 2.497