Table 1

Description of studies included in the meta-analysis

First author, year	Participants	Study topic	Interventions	Questionnaires returned/sent
Asch 1998	1000 primary care physicians identified through the American Medical Association Physician Master File (US).	Attitudes about cost containment in cancer screening.	1. $2 sent with questionnaire 2. $5 sent with questionnaire	221/500 (44.2%) 296/500 (59.2%)

Bellizzi 1986	200 people randomly selected from a city telephone directory, (US).	Shopping convenience and lifestyle.	1. No incentive 2. $1 bill included with questionnaire	37/100 (37%) 54/100 (54%)

Berk 1993	315 physicians (allergists and otolaryngologists) in Project HOPE survey (US).	Tests and costs for diagnosing allergies.	1. $10 with first mailing; follow up questionnaire and letter mentioning the incentive	79/125 (63.2%)
			2. No incentive with first mailing; Follow up questionnaire with $10 and letter explaining the importance of the study	62/125 (49.6%)
			3. No mention of $10 in either first or second mailing	26/65 (40.0%)

Biner 1988	200 of the 77216 residents of a mid-western city, randomly sampled from a telephone directory (US).	Community needs assessment.	2×2 factorial design: No incentive v $1	30/100 (30%) v
			Reactance appeal v No reactance appeal	61/100 (61%)

Biner 1990	200 of the 120500 residents of a mid-western city, systematically sampled from a telephone directory (US).	Community needs assessment.	2×2 factorial design: $1 v $0.25	67/100 (67%) v
			Obliged to reply v Given in appreciation	46/100 (46%)

Brennan 1991	900 people listed on one of the 57 electoral rolls representing the main urban centres (New Zealand).	Personal investments, banking and finance.	1. No incentive 2. 20c coin with first mailing	42/100 (42%) 46/100 (46%)
			3. 50c coin with first mailing	65/100 (65%)
			4. $1 with first mailing	49/100 (49%)
			5. 20c coin with second mailing	53/100 (53%)
			6. 50c coin with second mailing	45/100 (45%)
			7. $1 with second mailing	56/100 (56%)
			8. Entry into prize draw for $200 cash offered with each mail out	45/100 (45%)
			9. Entry into prize draw for $200 gift voucher offered with each mail out	50/100 (50%)

Brennan 1992a	400 people listed on a financial service company’s ‘hot prospect’ list (New Zealand).	Financial services.	1. No incentive 2. $0.50	144/200 (72%) 162/200 (81%)

Brennan 1992b	400 people listed on the electoral roll (New Zealand).	Use of brands of fruit juice.	1. No incentive	112/200 (56%)
			2. $0.50	128/200 (64%)

Brennan 1992c	384 people listed on the electoral roll (New Zealand).	Shopping.	1. No incentive	123/192 (64.1%)
			2. $0.50	134/192 (69.8%)

Brennan 1993a	412 dairy and beef farmers (New Zealand).	Use of mineral supplements.	1. No incentive	50/101 (49.5%)
			2. $0.50 coin with first mailing	70/103 (68.0%)
			3. $1 coin with first mailing	66/100 (66.0%)
			4. $1 lottery ticket with first mailing	64/108 (59.3%)

Brennan 1993b	1847 people listed on the electoral roll (New Zealand).	Attitudes toward social inequality.	1. No incentive	230/452 (50.9%)
			2. $0.50 coin with first mailing	278/454 (61.2%)
			3. $1 coin with first mailing	307/463 (66.3%)
			4. Promise that $1 would be donated to a charity for each valid return (in each of three mailings)	271/478 (56.7%)

Burns 1980	400 bank, savings and loan chief executive officers (US).		2×2 factorial design:	54/200 (27.0%)
			No incentive v $0.25	v
			No follow up v Follow up postcard sent 10 days after initial mailing	89/200 (44.5%)

Camunas 1990	700 members of the New York State Nurses Association (US).	Nursing association activities.	1. No incentive 2. $1 bill	63/300 (21.0%) 90/200 (45.0%)
			3. Brochure	27/200 (13.5%)

Collins 2000	4280 members of the RAND adolescent and young adult panel study drawn from schools across the US (US).	Smoking and substance misuse.	1. $20 cash with mailing 2. $20 cash on return of questionnaire	1047/1689 (62.0%) 1019/1734 (58.8%)
			3. $25 cash on return of questionnaire	564/857 (65.8%)

Deehan 1997	3584 GPs who did not respond to two mailings of a questionnaire (UK).	Work with patients who misuse alcohol.	Third mailing: 1. No incentive	101/1188 (8.5%)
			2. £5 charity donation	57/607 (9.4%)
			3. £10 charity donation	58/578 (10.0%)
			4. £5	105/613 (17.1%)
			5. £10	135/598 (22.6%)
			Fourth mailing to non-respondents in control group of third mailing:
			1. £5	63/534 (11.8%)
			2. £10	85/536 (15.9%)

Denton 1988	297 graduates from the Department of Educational Curriculum and Instruction at a large university in the south west (US).	Educational knowledge and skills.	2×2 factorial design: No incentive v $0.25 No newsletter v newsletter	109/257 (42.4%) v 16/40 (40.0%)

Denton 1991	100 graduates from the Department of Educational Curriculum and Instruction at a large university in the south west: (US).	Educational knowledge and skills.	1. No incentive 2. $0.25 3. $0.50	12/20 (60.0%) 13/20 (65.0%) 14/20 (70.0%)
			4. $1	14/20 (70.0%)
			5. Raffle	13/19 (68.4%)

Dommeyer 1980	352 people listed in the Cincinnati telephone directory (US).	Attitudes to mailed surveys.	No incentive v pre-paid $0.25 in follow up	21/176 (11.9%) 30/176 (17.0%)

Dommeyer 1988	600 people listed in the Chicago and Pheonix telephone directory (US).	Morality and conscience.	1. No incentive 2. $0.25 coin	37/100 (37%) 50/100 (50%)
			3. $0.25 cheque	37/100 (37%)
			4. $0.25 money order	38/100 (38%)
			5. ‘Early bird’—get a share in an incentive ($25) if send questionnaire back quickly	33/100 (33%)
			6. Sweep stake (entered into sweepstake to win $25 if return questionnaire by deadline)	30/100 (30%)

Donaldson 1999	400 physicians randomly selected from a list of US physicians actively caring for at least one transplant patient (US).	Interest in an internet consulting service.	2×2 factorial design: No incentive v $5 cheque with initial mailing No follow up call v follow up call to non-respondents 4 weeks after initial mailing	91/200 (45.5%) v 115/200 (57.5%)

Doob 1971a	400 people listed in a phone book (Toronto and Ontario, Canada).	Lifestyle.	2×2 factorial design: No incentive v dime incentive	60/200 (30.0%) v
			No reactance v reactance (‘reactance’ was a request written to make participants feel an attempt was being made to limit their freedom)	93/200 (46.5%)

Doob 1971b	200 people listed in a phone book (Toronto and Ontario, Canada).	Lifestyle.	2×2 factorial design: No incentive v dime incentive	30/100 (30%) v
			No reactance v reactance	48/100 (48%)

Doob 1971c	200 people listed in a phone book (Toronto and Ontario, Canada).	Lifestyle.	2×2 factorial design: No incentive v $0.20 incentive	36/100 (36%) v
			No reactance v reactance	53/100 (53%)

Doob 1973	804 people listed in a California telephone directory (US).	General.	1. No incentive 2. $0.05	110/268 (41.0%) 159/268 (59.3%)
			3. $0.20	191/268 (71.3%)

Doody 2003a	1500 radiologist technologists who had not responded to two earlier mailings of a questionnaire (US).	Physician diagnosed cancers and risk factors.	1. No incentive 2. $1 bill	46/300 (15.3%) 74/300 (24.7%)
			3. $2 bill	86/300 (28.7%)
			4. $2 cheque	63/300 (21.0%)
			5. $5 cheque	82/300 (27.3%)

Doody 2003b	1200 radiologist technologists who had not responded to two earlier mailings of a questionnaire (US).	Physician diagnosed cancers and risk factors.	1. No incentive 2. $1 bill	53/300 (17.7%) 71/300 (23.7%)
			3. $2 bill	84/300 (28.0%)
			4. $2 cheque	66/300 (22.0%)

Fiset 1994	517 dentists insured by a major malpractice carrier in two western states (US).	Malpractice claims experience.	1. $5 2. $10	156/258 (60.5%) 163/259 (62.9%)

Friedman 1979	600 people listed in telephone directories of two suburban areas in the greater New York Metropolitan area (US).	Public attitudes toward ethnic groups.	2×2 factorial design: No incentive v $0.25 Sponsor name suggests black race v white race	66/300 (22.0%) v 113/300 (37.7%)

Furse 1982	600 microwave oven owners listed in a major manufacturer’s warranty registration records (US).	Microwave ovens.	3×2 factorial design: No incentive v $0.50 enclosed v $1 enclosed	110/200 (55.0%) v
			No charity incentive v charity incentive (promise of $1 to charity of respondent’s choice for returned questionnaire)	154/200 (77.0%)

Gajraj 1990	700 customers of a major public utility, comprising households in south western Ontario (Canada).	Energy conservation.	1. No incentive 2. $0.50 included	34/100 (34%) 62/100 (62%)
			3. Promise of $0.50 on return of completed questionnaire	42/100 (42%)
			4. Pen included	41/100 (41%)
			5. Promise of pen on return of completed questionnaire	43/100 (43%)
			6. Inclusion in share of winning from five Super Lotto lottery tickets	55/100 (55%)
			7. Promise of inclusion in share of five Super Lotto lottery tickets on return of completed questionnaire	45/100 (45%)

Gendall 1998	1890 people on the electoral roll (New Zealand).	Role of government and attitudes to work orientation.	1. No incentive 2. Foil wrapped tea bag included	386/632 (61.1%) 388/631 (61.5%)
			3. $1 coin included	431/627 (68.7%)

Gibson 1999	1446 residents of Pierce County who had been clients of Medicaid and 797 residents who had been clients of Washington Basic Health Plan for more than 6 months (US).	Health and use of health services.	1. No incentive 2. $1 3. $2	209/412 (50.7%) 874/1467 (59.6%) 139/364 (38.2%)

Gillpatrick 1994	619 engineers identified from the subscriber list of a major trade journal (US).	Computer aided design workstations.	2×2 factorial design: No incentive v $1	41/213 (19.2%) v
			No pre-contact v pre-contact	178/406 (43.8%)

Glisan 1982	1512 farmers from six geographical regions (US).	Farm operations and crops.	3×3×2 factorial design:	73/504 (14.5%)
			No incentive v $0.25 v promised survey results	v
			Tan v blue v white questionnaire	122/504 (24.2%)
			Regular stamp v commemorative stamp	v
				98/504 (19.4%)
Godwin 1979	232 people in 60 countries, including government ministers, senior academics and directors of relevant organisations.	Facts and opinions on politics of family planning programs.	1. No incentive 2. $25 on return of questionnaire 3. $50 on return of questionnaire	46/72 (63.9%) 57/77 (74.0%) 62/83 (74.7%)

Goodstadt 1977	2416 readers of Addictions magazine (US).	Evaluation of Addictions magazine.	1. No incentive	375/604 (62.1%)
			2. $0.25	451/604 (74.7%)
			3. Free book incentive	403/604 (66.7%)
			4. Promise of free book on return of completed questionnaire	396/604 (65.6%)

Groeneman 1986	600 people listed in the telephone directories of Toronto, Montreal, Vancouver and Winnipeg selected using ‘distinctive Jewish name sampling’ (Canada).	Travel experiences and attitudes toward possible future trips.	1. No incentive 2. $1 enclosed	51/300 (17%) 102/300 (34%)
Hackler 1973	218 mothers of ninth or tenth grade students living in one neighbourhood of Edmonton (US).	Community cohesiveness.	1. No incentive 2. $1 bill incentive	43/109 (39.4%) 77/109 (70.6%)

Halpern 2002	1,200 general internists and family practitioners randomly selected from the American Medical Association’s master file of physicians (US).	The comparative merits of placebo controlled versus active controlled trials of novel antihypertensive drugs.	2×2×2 factorial design: $5 v $10 incentive Small v large outgoing envelope	354/700 (50.6%) v 293/500 (58.6%)
			Peppermint candy v none

Hancock 1940	6263 people listed on the personal tax records of the county assessors (US).	Attitudes toward chain and independent stores.	1. No incentive 2. $0.25 included	366/3726 (9.8%) 463/960 (48.2%)
			3. Promise of $0.25 on return of questionnaire	204/1123 (18.2%)
			4. Personal interview	388/454 (85.5%)

Hansen RA 1980	2496 industrial safety engineers employed by firms that require employees to wear safety hard hats (US).	Work related.	1. No incentive 2. $0.25	114/832 (13.7%) 308/832 (37.0%)
			3. Ballpoint pen worth $0.25	177/832 (21.3%)

Hopkins 1988	507 professional school and public librarians (US).	Attitudes and practices toward holding books in Spanish.	1. No incentive 2. $1 incentive	188/253 (74.3%) 218/254 (85.8%)

Hubbard 1988a	2,000 residents of a major mid-western metropolitan area (US).	Satisfaction with a range of banking and financial services.	1. No incentive 2. Promise of $1 donation to charity of respondent’s choice	162/400 (40.5%) 134/400 (33.5%)
			3. $0.25 enclosed	227/400 (56.8%)
			4. $1 enclosed	272/400 (68.0%)
			5. Opportunity to win $200 cash prize	207/400 (51.8%)
Hubbard 1988b	3150 residents of a major mid-western metropolitan area (US).	Satisfaction with a range of banking and financial services.	1. No incentive 2. Pre-paid $0.25	121/450 (26.9%) 182/450 (40.4%)
			3. Pre-paid $1	254/450 (56.4%)
			4. Opportunity to win $50	133/450 (29.6%)
			5. Opportunity to win $100	127/450 (28.2%)
			6. Opportunity to win $150	129/450 (28.7%)
			7. Opportunity to win $200	170/450 (37.8%)

Huck 1974	200 students living in residence halls at the university of Tennessee (US).	A variation of the Rokeach Dogmatism Scale.	1. First mailing with $0.25 2. Second mailing (to non-respondents) with $0.25	47/50 (94%) 46/50 (92%)
			3. Third mailing (to non-respondents) with $0.25	39/50 (78%)
			4. First, second and third mailings without $0.25	36/50 (72%)

James 1990	844 cable television subscribers in Fairfax County, Virginia (US).	Personal information.	1. No incentive 2. $0.25	148/168 (88.1%) 146/169 (86.4%)
			3. $0.5	145/168 (86.3%)
			4. $1	157/169 (92.9%)
			5. $2	163/170 (95.9%)

James 1992	1200 members of a national trade association of owners of construction subcontracting companies who were not currently enrolled in the association’s health insurance programme (US).	Personal information about employees.	1. No incentive 2. $1 cash 3. $5 cash 4. $5 cheque	78/150 (52.0%) 96/150 (64.0%) 107/150 (71.3%) 101/150 (67.3%)
			5. $10 cheque	100/150 (66.7%)
			6. $20 cheque	119/150 (79.3%)
			7. $40 cheque	104/150 (69.3%)
			8. $50 on return of completed questionnaire	85/150 (56.7%)

Jobber 1988	159 chief executives of building societies (UK).		2×2 factorial design:	23/79 (29.1%)
			No incentive v 20 pence	v
			No booklet v booklet explaining survey included	35/80 (43.8%)

John 1994	7549 cosmetologists aged 22–36 years (US).	Recent health problems, pregnancy and childbirth.	1. No incentive 2. $1 in first mailing	327/443 (73.8%) 2257/2791 (80.9%)
			3. $1 in second mailing	3385/4315 (78.4%)

Kasprzyk 2001	300 general internists listed on the American Medical Association files who spend time on direct patient care, dealing with STD diagnosis (US).	STD diagnosis, treatment and control practices.	3×2 factorial design: No incentive v $15 cash v $25 cash First class mailing v FedEx mailing	25/100 (25%) v 67/100 (67%)
				v
				66/100 (66%)

Keown 1985a	100 business executives (Japan).	Attitudes to business risks.	1. No incentive	11/50 (22%)
			2. $1	24/50 (48%)

Keown 1985b	100 business executives (Hong Kong).	Attitudes to business risks.	1. No incentive	6/50 (12%)
			2. $1	0/50 (0%)

Kephart 1958	500 women who had passed their Pennsylvania State Nursing Board exams (US).	Attitudes to aspects of the nursing profession.	1. No incentive 2. Penny	52/100 (52%) 55/100 (55%)
			3. Nickel	54/100 (54%)
			4. Dime	57/100 (57%)
			5. Quarter	70/100 (70%)
Leung 2002	4850 physicians randomly selected from the full and limited registration lists of the Hong Kong Medical Council (Hong Kong).	Computerisation of clinical and administrative tasks among physicians.	1. No incentive 2. HK$10 cash 3. HK$20 cash	281/1700 (16.5%) 9/50 (18.0%) 15/50 (30.0%)
			4. HK$40 cash	17/50 (34.0%)
			5. Entry into HK$1000 lottery	175/1000 (17.5%)
			6. Entry into HK$2000 lottery	197/1000 (19.7%)
			7. Entry into HK$4000 lottery	203/1000 (20.3%)

London 1990	1000 electronics design engineers (US).	Use of components.	1. No incentive	24/500 (4.8%)
			2. $1	116/500 (23.2%)

Martinson 2000	4200 adolescents aged 14–17 years in the Minneapolis/St Paul metropolitan area (US).	Behaviour and attitudes toward smoking.	1. No incentive 2. $2 included with questionnaire	483/1050 (46.0%) 650/1050 (61.9%)
			3. $15 on completion and return of questionnaire	721/1050 (68.7%)
			4. Promise of entry into ten drawings for $200 cash prizes on completion and return of questionnaire	589/1050 (56.1%)

McConochie 1985	11057 men aged 18–34 (US).	Radio listening.	1. $0.50	2312/5249 (44.0%)
			2. $2	1448/2874 (50.4%)
			3. $5	1660/2934 (56.6%)

Mizes 1984	200 physicians specialising in allergy, randomly selected from all physicians listed under allergy or allergy/immunology in the telephone directory yellow pages of major metropolitan areas (US).	The incidence, treatment and success of treatment of rhinitis medicamentosa.	1. No incentive; Answer sheet on postcard 2. $1 cheque; answer sheet on postcard 3. $5 cheque; answer sheet on postcard 4. $1 cheque; answer on bank cheque	21/40 (52.5%) 29/40 (72.5%) 29/40 (72.5%) 21/40 (52.5%)
			5. $5 cheque; answer on bank cheque	29/40 (72.5%)

Paolillo 1984	400 business professionals from the mid-west (US).		1. No incentive	36/100 (36.0%)
			2. $1 enclosed with questionnaire	65/100 (65.0%)
			3. $2 promised on return of questionnaire	41/100 (41.0%)
			4. Entry into a lottery for a cash prize promised on return of questionnaire	33/100 (33.0%)

Parkes 2000	2561 adults aged 20–74 years who were controls in a Canadian case-control study of cancer (Canada).	Lifetime residential and occupational history, smoking, diet, physical activity and use of medication.	1. No incentive 2. $2 sent with questionnaire 3. $5 sent with questionnaire	519/838 (61.9%) 623/856 (72.8%) 669/867 (77.2%)

Peck 1981	5850 men and women who had been high school juniors in spring 1973 (US).	Work experience and career plans.	1. No incentive 2. Prepaid $3	566/1462 (38.7%) 1882/2918 (64.5%)
			3. $3 on return of questionnaire	773/1470 (52.6%)

Perneger 1993	1235 young adults enrolled in various insurance plans (Geneva, Switzerland).	Health status.	1. No incentive 2. Reminder card	243/311 (78.1%) 252/309 (81.6)
			3. Promise of 10 Swiss francs	260/310 (83.9%)
			4. Reminder card and promise of 10 Swiss francs	252/305 (82.6%)

Pressley 1977	280 marketing research directors (US).		2×2×4 factorial design:	45/140 (32.1%)
			No incentive v Dime	v
			Cartoons v No cartoons	61/140 (43.6%)
			Yellow v Blue v Green v White questionnaire

Roberts 2000	1000 English women aged 40–65 years (UK).	Menopause services in the north west of England.	2×2 factorial design: No incentive v payment of £5 on response	420/750 (56.0%) v
			No lottery incentive v entry into lottery for £50 on response	169/250 (67.6%)
Robertson 1978	450 people systematically selected from the Denver Metropolitan area phone directory (US).		1. No incentive 2. Promise of $1 cash on return of questionnaire	35/150 (23.3%) 39/150 (26.0%)
			3. Promise of $1 donation to charity on return of questionnaire	62/150 (41.3%)

Shackleton 1980	85 partially sighted school leavers aged 17–20 years who had left schools for the visually handicapped during the previous academic year (UK).	Occupational experiences during the first year after leaving school.	2×2 factorial design: No incentive v £1 offered No examination v Previous examination	38/42 (90.5%) v 40/43 (93.0%)

Shaw 2001	1800 HealthSystem Minnesota enrolees aged 20–80 years (US).	Quality of life, anxiety, depression, morbidity and health care utilisation.	1. $2 2. $5	590/900 (65.6%) 649/900 (72.1%)

Skinner 1984	300 marketing professors (Canada).	Needs of instructors for an introductory marketing text.	1. No incentive 2. $1 pre-paid	23/60 (38.3%) 30/60 (50.0%)
			3. $1 on return of completed questionnaire; Respondent identified	20/60 (33.3%)
			4. $1 on return of completed questionnaire; respondent not identified	15/60 (25.0%)
			5. $1 promised to charity	16/60 (26.7%)

Spry 1989	3388 residences listed in the Haines Directory who had not responded to a questionnaire (San Diego, US).	Health and physical activity.	1. No incentive 2. Promise of $5 on return of completed questionnaire	323/3114 (10.4%) 21/96 (21.9%)
			3. Promise of $1 on return of completed questionnaire	12/90 (13.3%)
			4. $1 bill enclosed	11/88 (12.5%)

Tullar 1979	200 large manufacturing firms (US).	Product development.	2×2 factorial design:	50/100 (50.0%)
			No incentive v $0.10	v
			No follow-up v Follow-up	78/100 (78.0%)

VanGeest 2001	873 physicians randomly selected from the American Medical Association’s master file of all practising physicians (US).	Attitudes toward, and responses to, utilisation review pressures.	1. $5 cash 2. $10 cash 3. $20 cash	176/292 (60.3%) 198/291 (68.0%) 189/290 (65.2%)

Warriner 1996	3044 households listed in the Grand River Watershed region of south-western Ontario (Canada).	Environmental issues.	4×5 factorial design: No monetary incentive v Canadian $2 v $5 v $10	219/347 (63.1%) v
			No offer to make a charitable donation v $2 offer v $5 offer v $10 offer v Lottery (five prizes of Can$200)	1441/2007 (71.8%) v 270/351 (76.9%)
				v
				263/339 (77.6%)

Weltzien 1986	942 people who had terminated from mental health treatment centres (US).	Client satisfaction.	1. No incentive 2. $0.02	86/471 (18.3%) 115/471 (24.4%)

Wiseman 1973	464 residents in a state-wide telephone listings, Massachusetts (US).	Attitudes and opinions toward the Massachusetts State lottery.	2×2×2×2 factorial design No incentive v $0.10	90/232 (38.8%) v
			No follow-up v Postcard follow-up 3 days after initial mailing	109/232 (47.0%)
			Stamped v Business reply return envelope
			No offer of survey results v offer of survey results
Wotruba 1966	150 urban household residents (US).		1. No incentive	9/50 (18%)
			2. $0.25 with questionnaire	20/50 (40%)
			3. $0.50 promised on return of questionnaire	10/50 (20%)

Zusman 1987	371 undergraduate transfer students (US).		1. No incentive	74/171 (43.3%)
			2. $1 sent with first mailing	122/200 (61.0%)