Annesi (2002) Italy | N = 100 [gym members] Intervention condition N = 50 Control condition N = 50 | All participants were told that completing three sessions (or more) of vigorous exercise per week was recommended for fitness progress. All participants were provided individual appointments of 40 min with the same exercise professional every 6 weeks Control condition’s meetings focused on the transfer of physiological knowledge, the need to continue exercise having positive effects on health and personalised modification of exercise plans on progress Intervention condition had an additional focus which was the implementation of a goal-setting protocol | Attendance was calculated for the 52 weeks of the study | Over the study period there was greater attendance in the intervention condition than the control condition (p < .0001) | 1.37 |
Calzolari et al. (2017) Italy | N = 247 [university students] Intervention condition N = 89 Control condition N = 158 | Intervention condition received weekly emails for reminding them of the opportunity to attend the gym (during a maximum period September 1, 2009 to March 16, 2010). Control condition did not receive reminders | Attendance was monitored during and two years after the treatment period | The intervention condition had 0.6 more visits per month than the control condition during the treatment period but this was not significant [p > 0.05] There was no significant difference in the number of visits per month between the intervention and control conditions at two year follow-up [p > 0.05] | 0.16 0.08 |
Carrera et al. (2017) USA | N = 690 [new gym members] Intervention condition N = 514 Control condition N = 176 | Intervention condition received one of three incentives if they attended the gym at least 9 times over the first 6 weeks of their membership; a 30 dollar payment (“money30”), or a 60 dollar payment (“money60”) or an item they had chosen costing 30 dollars (“item”) Control condition received 30 dollars payment unconditionally during the same period | Attendance was monitored for the first 12 weeks of the members’ gym membership (including the six week intervention period at the beginning of their membership) | For the intervention condition as a whole, incentives did not have a statistically significant impact on attendance during the first six weeks [p > 0.05 0.10] | 0.08 |
Carrera et al (2018) USA | N = 877 [members of a private gym] Intervention condition N = 438 Control condition N = 439 | Intervention condition were asked to check off the time they planned to work out that day each day in a 13 day period [participants were told that the information would be used to create calendar invitations for each day/time they planned to visit] Control condition were asked to check off a time that they worked out in the preceding two weeks | Attendance was monitored between the two conditions during the experimental period | There was no significant difference between the intervention and control conditions during the experimental period [p > 0.05] | 0.10 |
Courneya et al (1997) Canada | N = 300 [alumni, support staff, academic staff, and the general public. University students were excluded] Intervention condition N = 100 Active control condition N = 100 Control condition N = 100 | Intervention condition participants received a letter by mail, containing a friendly message and pamphlet outlining the possible activities available at the fitness facility. The letter included an additional paragraph instructing them that they could earn one month’s free membership if they attended the fitness facility at least 12 times in the next month Active control condition participants received the same letter by mail as the intervention condition, without the additional paragraph instructing them that they could earn 1 month’s free membership if they attended the fitness facility at least 12 times in the next month The control condition participants received no intervention | Attendance of all participants was monitored for one month following the intervention | The intervention condition had significantly higher attendance than the active control condition over the one month period [t(198) = 2.76, p < 0.05] The intervention condition did not have significantly higher attendance than the control condition [p > 0.05] | 0.38 0.14 |
Estabrooks et al (1996) Canada | N = 200 [Alumni, support staff, academic staff, and the general public. University students were excluded] Intervention condition N = 100 Active control condition N = 50 Control condition N = 50 | Intervention condition participants received a letter by mail, which contained a friendly message and outlined the possible activities at the fitness facility. They also received a key chain that was to act as a stimulus control and a brief statement about the purpose of the key chain. At the completion of the 8 week observation period, they received a telephone call as a manipulation check Active control condition received a letter by mail, which contained a friendly message and outlined the possible activities at the fitness facility. They did not receive the additional stimulus The control condition participants received no intervention | Attendance was monitored for eight weeks following the intervention | There was no main effect for the intervention condition [F (197) = .47, p > 0.05] | 0.05 |
Marz (2017) Germany | N = 94 [registered members of the gym] Intervention condition N = 60 Control condition N = 34 | Intervention condition participants were split into a “gain-treatment” or “loss-treatment”. In the “gain-treatment”, participants were rewarded for frequent attendance at the gym. In the “loss-treatment”, incentives were framed in a way that infrequent attendance at the gym was penalized Control condition participants received no financial incentives | Attendance was monitored for the four week intervention and 12 weeks after the intervention | Participants assigned to the “loss- treatment” had an estimated average of 0.686 additional visits per week in the intervention period compared to the control condition, which was statistically significant [p < 0.05] Participants assigned to the “gain-treatment” had an estimated average of 0.344 additional visits per week compared to the control condition, which was not statistically significant [p > 0.05] | d = 0.33 d = 0.23 |
Muller and Habla (2018) Sweden | N = 2463 [new registered members of the gym] Intervention condition N = 1231 Control condition N = 1232 | Intervention condition received a series of email reminders over the course of a 3 month period [January 9, 2017 and April 9, 2017] encouraging them to attend the gym. The control condition received no email reminders | Attendance data was analysed during the intervention period | During the intervention period, the intervention condition had a slightly higher attendance than the control condition [total visits increase by 13%] (p < 0.01) | d = 0.01 |
Nigg et al (1997)a Canada | N = 204 [Alumni, support staff, academic staff, and the general public] Intervention condition N = 154 Control condition N = 50 | The three experimental conditions received a letter by mail that contained a friendly message and a calendar month with large squares containing four weeks beginning November 13 and ending December 10. Participants were unaware that the study focused on the motivational effects of self-monitoring or that their attendance was being objectively monitored by the researchers. Participants in the "Positive SM" condition were instructed to place an "X" in each calendar day they attended the fitness facility. Participants in the "Negative SM" condition were asked to place an "X" in each calendar day that they did not attend the fitness facility. Participants in the "Neutral SM" were instructed to place a "tick" in each calendar day that they attended the facility and a "X" in each day they did not attend the facility. The control condition received no intervention | Attendance was monitored before the intervention and for four weeks post intervention | The “Positive SM” condition showed a significantly higher attendance than the control condition post-intervention (p < 0.05) The “Negative SM” condition showed a significantly higher attendance than the control condition post-intervention ( p < 0.05) The “Neutral SM” condition showed a non-significant difference in attendance post intervention compared to the control condition (p > 0.05) | d = 0.08 d = 0.20 d = 0.02 |
Nigg et al (1997)b Canada | N = 153 [Alumni, support staff, academic staff, and the general public] Intervention condition N = 102 Control condition N = 51 | Intervention condition participants received a telephone call ‘interview’ in which they were asked to think systematically of and record the expected gains and losses of either exercising at the gym (relevant scenario) or not smoking (irrelevant scenario) Control condition participants received no intervention | Attendance was monitored for four weeks of baseline and the eight weeks of the intervention The number and importance of pros and cons listed by each individual in the relevant DBS condition was examined | Attendance in the relevant DBS condition saw virtually no change from baseline to the end of the intervention [t(50) = .26, p > 0.05] while attendance in the control condition saw a significant decrease from baseline to the end of the intervention f(50) = 1.94. p < .03. | d = 0.31 |
Rohde et al. (2017) Netherlands | N = 1182 [members of the gym] Intervention condition N = 258 (Unconditional rebate n = 48; Conditional rebate n = 113; Choice n = 97) Control condition N = 924 | Intervention condition participants were randomly split into ‘conditional’, ‘unconditional’ or ‘choice’ conditions. The ‘conditional’ participants received a rebate of approximately 10% of the average membership fee conditional on attending the gym at least once per week in 11 of the 13 weeks of the first quarter in 2010. This incentive was repeated in the following quarter. The ‘unconditional’ condition participants received the 10% rebate per quarter for staying a member of the gym. The ‘choice’ participants could choose between the conditional or unconditional rebate. Control condition participants did not receive any incentives | Attendance of participants was monitored for 15 months in total; the quarter before the interventioz, the two quarters of the intervention and the two quarters following the intervention | The only increase in attendances during the intervention period was for the conditional rebate (CR) and unconditional rebate (UR) conditions in the first quarter of 2010. There was no effect when comparing each of the intervention conditions to the control condition (p > 0.05). | [UR: d = -0.03] [CR: d = -0.004] |
Spangenberg (1997) USA | N = 142 [members of the club] Intervention condition N = 73 Control condition N = 69 | Intervention condition participants received a telephone call asking whether they were a member of a health club and then asked “Do you expect to use the club in the next week?” Control condition participants received the same telephone call as the intervention condition, but were not asked the question “Do you expect to use the club in the next week?” | Attendance was monitored for the 10 day period immediately following telephone contact and for the six-month period following the intervention | Over the ten day period, 12% of the intervention condition participants and 7% of the control condition participants attended the club once or more during the ten day period, however this was not statistically significant (χ2 = 1.12, df = 1, p > 0.05) For the six month period, the average number of visits was 10.25 for the intervention condition which was double the control condition average of 5.1 visits. This was significant at the 5% confidence level (F(l,93) = 3.78, p = 0.05). | d = 0.18 d = 0.10 |
Thompson et al (1980) Canada | N = 36 [adult female members of the gym] Intervention condition N = 18 Control condition N = 18 | All participants were contacted by telephone to arrange a meeting for a new exercise programme offered by the club. Participants at this initial meeting were asked to complete a series of personal inventories and to express their relative preferences for a number of exercises Participants were then randomly assigned to the treatment conditions and returned for a second visit to the club Intervention condition participants were told that their programme was based totally on the choices they had made. At the end of the second visit they were asked to select six additional exercises which they would add to their programme – one every third visit Control condition participants were told that their programme was based on a standardised exercise format rather than on their expressed preferences. At the end of their second visit they were told that six additional exercises would be added to their programmes by the instructors | Attendance was monitored over a six week period following the intervention | The intervention condition had a higher average attendance than the control condition over the 6-week period, however, this was not statistically significant [F(1,34) = 2.88, p > 0.05] | d = 0.29 |
Wankel et al (1977) Canada | N = 100 [adult female members of the gym] Intervention condition = 75 [‘Complete decision’ n = 25, ‘Positive-only’ n = 25, ‘Regular call up’ n = 25, Control condition n = 25 | The ‘complete decision-balance-sheet’ treatment received a telephone call where they were asked to complete a decision balance sheet grid concerning attendance of the health club's programmes. The ‘positive-only’ telephone interview condition were only asked to think of and report positive outcomes to be expected. A further condition (‘Regular call up ‘received a standard telephone call utilised by the club in following up inactive members. This call attempted to establish why members had not been utilising their membership and encouraged them to use it more in the future. This condition served as a "personal attention" control condition for the other two intervention conditions The control condition received no intervention | Attendance was monitored for one month following the intervention | The three treatment conditions had a significantly higher attendance than the control condition (p < 0.05) The “Positive only” condition had the highest attendance compared to the control condition. This was followed by the ‘complete decision-balance-sheet’ condition. The regular call up condition had the smallest attendance difference from the control group | d = 1.45 d = 1.00 d = 0.60 |