Table 2

Response rates overall and by design characteristics, by gender, race, and age

2 × 2 Factorial Design


Overall

Pages

Notification


2

4

Letter

Postcard



Overall

363 (46.5)

169 (44.6)

194 (48.4)

179 (46.3)

184 (46.8)



Gender


Female

212 (54.1)

105 (54.4)

107 (53.8)

108 (54.8)

104 (53.3)


Male

151 (38.9)

64 (34.4)

87 (43.1)

71 (37.4)

80 (40.4)


P value†

< 0.0001

0.007

0.03

0.0006

0.01



Race


White

306 (52.6)

147 (50.5)

159 (54.6)

151 (52.6)

155 (52.5)


Non-white *

14 (24.14)

6 (22.2)

8 (25.8)

5 (15.6)

9 (34.6)


Unknown/Other

43 (30.7)

16 (26.2)

27 (34.2)

23 (33.8)

20 (27.8)


P value†

< 0.0001

< 0.0001

0.0002

< 0.0001

0.0003



Age


25 - 29

35 (30.7)

12 (21.8)

23 (39.0)

12 (25.0)

23 (34.8)


30 - 39

81 (40.7)

32 (32.3)

49 (49.0)

42 (41.2)

39 (40.2)


40 - 49

107 (50.7)

51 (52.0)

56 (49.6)

57 (51.8)

50 (49.5)


50 - 59

103 (55.1)

56 (59.0)

47 (51.1)

50 (54.4)

53 (55.8)


60 - 65

37 (53.6)

18 (56.2)

19 (51.4)

18 (51.4)

19 (55.9)


P value†

0.0002

< 0.0001

0.63

0.008

0.04


†P-values comparing response rates within condition from chi-square test or Fisher's exact test where appropriate.

Note: In the factorial design, response rates did not differ significantly by survey length (p = 0.29) or by notification type (p = 0.87). Combining pages and notification into 4 separate groups, the p-value for the difference in response rate was 0.56 (results not shown).

* "Non-white" includes: American Indian/Alaskan Native, Black or African American, Native Hawaiian/Pacific Islander, and Asian.

Beebe et al. BMC Medical Research Methodology 2010 10:50   doi:10.1186/1471-2288-10-50

Open Data