The impact of self-interviews on response patterns for sensitive topics: a randomized trial of electronic delivery methods for a sexual behaviour questionnaire in rural South Africa

Table 3 Item response rates for sexual behaviour questions not previously used in the surveillance

	ITT analysis							AT analysis
	PI arms		SI arms		Z	p	φ	PI arms		SI arms		Z	p	φ
Total number of respondents	174		166					203		137
Given gifts in past 12 months
Yes	79	45%	67	40%	0.46	0.50	0.04	95	47%	51	37%	2.04	0.15	0.08
Declined to answer	8	5%	13	8%	1.53	0.22	0.07	9	4%	12	9%	2.64	0.10	0.09
Given gifts to get sex in past 12 months ^a
Yes	10	13%	12	15%	0.85	0.36	0.05	12	13%	10	20%	1.20	0.27	0.06
Declined to answer	0	0%	1	1%	1.19	0.28	0.06	1	0%	0	0%	0.54	0.46	0.04
Received support in past 12 months
Yes	60	34%	63	34%	1.11	0.29	0.06	69	34%	54	39%	2.23	0.13	0.08
Declined to answer	6	3%	14	8%	3.81	0.05	0.11	7	3%	13	9%	5.39	0.02	0.13
Had sex to get support in past 12 months ^a
Yes	1	2%	3	5%	0.97	0.33	0.05	1	1%	3	6%	1.68	0.20	0.07
Declined to answer	0	0%	1	1%	0.96	0.33	0.05	0	0%	1	1%	1.29	0.26	0.06
Ever had anal sex
Yes	5	3%	6	3%	0.18	0.67	0.02	5	2%	6	4%	0.95	0.33	0.05
Declined to answer	10	6%	13	8%	0.59	0.44	0.04	14	7%	9	7%	0.01	0.91	0.01
Ever had sexual experience with same gender
Yes	2	1%	5	3%	1.55	0.21	0.07	4	2%	3	2%	0.03	0.86	0.01
Declined to answer	3	2%	7	4%	1.85	0.17	0.07	4	2%	6	4%	1.66	0.20	0.07
Ever had someone try to make you have sex against your will
Yes	7	4%	10	5%	0.88	0.35	0.05	10	5%	7	5%	0.03	0.87	0.01
Declined to answer	0	0%	7	4%	7.49	0.01	0.15	1	0%	6	4%	6.13	0.01	0.13

ITT Intention-to-treat, AT As-treated, PI personal interview arms (PAPI, CAPI), SI self-interview arms (CASI, ACASI), IQR inter-quartile range. Z test statistics compare PI to SI arms. In each pair the upper value is a comparison of the proportion of affirmative responses amongst valid responses and the lower value is a comparison of item non-response rates using a \( {\chi}_1^2 \) test. φ is the effect size associated with the relationship between interview arm and the outcome of interest. ^a These questions were only asked of those responding “Yes” to the preceding question. None of the questions in this table had previously been asked in annual surveillance questionnaires

ISSN: 1471-2288