market research resources for online surveys

professional marketing associations & paid survey resources

AACR - Association for Consumer Research
AAPOR - American Association for Public Opinion Research
AMA - American Marketing Association
ARF - Advertising Research Foundation
CASRO - Council of American Survey Research Organizations
QRCA - Qualitative Research Consultants Association
EFAMRO - European Federation of Market Research Agency Associations
ESOMAR - European Society for Opinion and Marketing Research
MRA - Marketing Research Association
MRS - Market Research Society United Kingdom
MRSA - Australian Market and Social Research Society
NCPP - National Council on Public Polls
AIMRI - Alliance of International Market Research Institutes
PMRG - Pharmaceutical Marketing Research Group
DMA - Direct Marketing Association

Check out these web surveys sites, and try out their search function (put in the word "cash" or "incentive") for thousands of articles and white-papers on all of this online surveys for cash stuff.

WebSM - Web Survey Methodology

NBRI - National Business Research Institute one of the many articles backing up the incentive surveys out there

IJPOR - International Journal of Public Opinion Research (not free access anymore, but click on the "abstract" link and you'll get the idea) -  just one of the many articles on this paid surveys online stuff...

Incentives in Private Sector Surveys

OK, this is just ONE sample of abstracts for presentations given at ONE seminar - the 2008 American Association for Public Opinion Research 63rd Annual Conference. I found this on Web Survey Methodology. Get the idea. The paid online surveys thing is a legitimate way to make money online - they are NOT scams. There are people that do nothing but study the effects of incentives on opinion surveys. Crazy, but true. If this doesn't prove it for you, not sure what will. If you do a search, you can find the more recent conferences as well.



Arbitron, Inc.

Surveys conducted by private sector organizations and commercial establishments face response rate challenges similar to those conducted by government agencies or academic institutions. Respondent cooperation can be influenced by such attributes as topic salience, survey sponsorship, data collection mode, and incentives, among others. These factors are often closely associated with the survey organization itself, and therefore, different types of survey organizations may vary in their opportunity to manipulate these attributes to effect higher response rates. This paper will discuss the use of incentives among survey organizations in the private sector, including how the relative importance of incentives compared to other survey attributes differs from surveys in other sectors. I will provide an overview of current practices in private sector surveys and the decision process to use incentives, including cost-benefit analysis. Many private sector organizations have tested a wide variety of cash and non-cash incentives. While cash consistently outperforms non-cash incentives, rising survey costs to maintain response rates have led to recent reconsideration and testing of non-cash incentives. Current practices with non-cash incentives will be presented. The private sector incentive practices provide an interesting comparison to incentive use in government and other survey organizations. While our overall incentive levels are comparable across the sectors, the processes for implementing incentives differ. This presentation will help in better understanding the commonalities and differences in use of incentives by type of survey organization.
 

Use of Incentives in Federal Surveys

U.S. Census Bureau  U.S. Office of Management and Budget

Incentives for survey respondents have been used extensively for many years to improve response rates. There is considerable research evidence supporting the value of respondent incentives for increasing cooperation and improving the rates, speed and quality of response in a broad range of data collection efforts. Over time, it has become more difficult to achieve high response rates in many key surveys conducted by the Federal government and academia. The use of respondent incentives as one tool to combat declining response rates (and addressing the potentially greater threat of non-response bias) has become increasingly widespread. This paper will provide a current overview of how incentives are being used across a wide range of surveys conducted by Federal agencies. Because all Federal agencies are required by the Paperwork Reduction Act to obtain approval from the Office of Management and Budget (OMB) prior to conducting a survey, we used OMB’s database to select all information collections that were submitted for review and approved between September 1, 2006, and December 31, 2007, that also designated use of surveys, censuses or statistical methods. From the 454 information collections meeting these criteria, we identified those that indicated use of respondent incentives, and coded several attributes of the studies, including the type of study, its primary purpose, average annual burden, and target population, along with the incentive type, timing and amount. We will present a summary of this information, providing a snapshot of how incentives are currently being used across Federal agencies.
 

How Much Is It Worth to You? A Meta-Analysis of the Impact of Incentives on Study Participation

NORC/University of Chicago

In recent years, numerous surveys have begun offering incentives in an attempt to improve study participation rates. These surveys have generated a large number of research reports on the impact of incentives on study participation in general and response rates in particular. With some exceptions, these studies have found incentives increase response rates without adding bias to survey estimates. Further, studies have generally found that larger incentive amounts are more effective than smaller incentive amounts, but that the benefit of increasing the incentive amount diminishes as the overall size of the incentive amount increases. These studies have been valuable in demonstrating the general utility of incentives for improving response rates but what is required now is a synthesis of the existing research into a theoretical framework. Such a framework should allow future studies to utilize incentives (and determine the appropriate amount of the incentive) without first requiring testing of the effectiveness on a particular study population. This paper proposes such a theoretical framework. We ground our theory in a meta-analysis of incentive studies since 2000. By combining these disparate studies via a meta-analytic framework, we are able to predict the impact of incentives on disparate populations and estimate an “incentive function” that predicts the impact on response rates of each dollar increase in the incentive amount. We discuss how researchers can apply our results to their studies and conclude with a discussion of the limitations of our results.

The Role of Cash Incentives in Online Panelist Motivations: Experimental Results on Unit Response and Satisficing

Western Wats, Inc.

Online survey panelists are frequently characterized as “professional respondents” who’s survey participation is motivated by self-interested incentive-maximization. Little scientific evidence exists to evaluate such claims. This paper reports on two experiments testing hypotheses at the intersection of survey incentives and unit-response; hypotheses that contrast theories of “economic exchange” and “social exchange.” A random sample of online survey panelists were randomly assigned to conditions varying both cash incentive and length of survey. Participants did not systematically respond more, or satisfice less when promised a higher cash incentive. Nor did participants respond more to a prepaid versus promised incentive. The results provide little support for a one-dimensional–“professional respondents”–characterization of online panelist motivations.
 
An Examination of the Best Method for Leveraging Large Cash Incentives, Nielsen Media Research

Nielsen Media Research

As incentives have become a routine part of many data collection efforts, understanding how best to leverage these incentives – particularly larger cash incentives – is crucial for maximizing participation while minimizing costs. In an earlier study, Trussell et al. (2006) reported that $30 cash was the most cost-effective incentive amount to use with hard-to-reach households recruited for Nielsen’s TV Ratings diary survey. We sought to determine how best to leverage this $30 cash incentive during the diary placement call. Households are sampled using random digit dialed methodology, contacted by telephone and asked to participate in the TV Ratings. Those who agree are sent a paper diary and an incentive. We attempted to leverage the incentive with hard-to-reach demographic groups using three strategies. First, we investigated whether this larger incentive amount would be construed as an economic exchange rather than a social exchange (Dillman 1978). If the incentive is considered to be an economic exchange, then we should refer to the incentive as a “payment” rather than as a “thank-you”. Second, we examined where in the interview to best mention the incentive – at the end, once the household agreed to keep the diary, or at the outset of the diary placement call. Third, we looked at whether sending different denominations (three $10 bills or one $10 and one $20 bill) made a difference on final response rates. Using a 2x2x2 design with a control group, approximately 75,000 households were included in the study. Analysis showed that emphasizing the $30 incentive early in the call was the most effective method for leveraging the incentive. In contrast, referring to the incentive as a “payment” or using different denominations of bills did not lead to significant gains in cooperation. These findings are discussed in relation to incentive leveraging and social exchange theory.

Respondent Incentives: Do They Alter Data Quality or Study Findings?

NORC/University of Chicago & National Center for Health Statistics, CDC

As response rates for Random Digit Dial (RDD) household surveys have declined over the past decade, the use of monetary incentives has increased. While this technique has generally improved response rates and coverage, the impact on overall data quality and survey findings is unclear. A common assumption is that incentive use should increase the interview completion rate and improve data quality – but is this true? Data from the 2005-2006 National Survey of Children with Special Health Care Needs will be examined. This RDD survey, completed in early 2007, was conducted by the State and Local Area Integrated Telephone Survey (SLAITS) mechanism of the National Center for Health Statistics on behalf of the Maternal and Child Health Bureau. These data can be used to produce prevalence estimates of children with special health care needs (CSHCN), and describe the types of services these children need and use. Detailed data were collected from a knowledgeable parent or guardian for approximately 43,000 CSHCN under 18 years of age. Fifteen-dollar ($15) incentives ($5 prepaid, $10 upon completion) were offered to parents or guardians who had not yet completed the interview. Our analyses will focus on whether incentive cases differed from non-incentive cases, and if so, how. We will also explore the impact of incentive cases on data quality and survey findings. Items to be examined include, but are not limited to, household education level, child’s gender, race of child or parents, total household income, geography, special health care need type(s), and household composition.

Effects of Incentives, Advance Letters, and Telephone Follow-up in RDD Recruitment for a Consumer Research Panel

Gallup Organization

Survey Research and Methodology, University of Nebraska Lincoln

Survey response rates have been declining for past several years, particularly for random digit dial (RDD) telephone surveys (see De Leeuw & De Heer, 2002; Steeh 1982). This trend affects consumer research panels like The Gallup Panel that uses Random Digit Dialing (RDD) to recruit its members. If significant improvements in panel recruitment response rates are to be achieved, new approaches must be considered. This paper presents the findings of an experiment conducted by The Gallup Panel to analyze the individual and combined effect of incentives, advance letters, and telephone follow-up on panel recruitment response rate. The experiment consisted of eight conditions in addition to the control group that represented the existing recruitment methodology, which involved drawing a RDD sample, recruiting households on the phone, sending a welcome packet questionnaire in mail, and following-up for non-response after twenty one days. Preliminary results suggest that response rate varies across conditions and advance letter with incentive and telephone follow up received the highest panel recruitment response rage. This article concludes with recommendations and future research avenues.

The Effect of Promised Incentives on the Response Rates of a Mail Survey

Arbitron, Inc.

With growing concerns about the viability of RDD frames, Arbitron began investigating the efficacy of address-based sample frames and the use of mail surveys without phone contact. As part of this study, we examined the effect of promised incentives ranging from $5 to $30 on the response rate of a two wave mail survey. A fifteen-question survey was sent to nearly 7,100 households whose addresses were unable to be matched to a phone number. The questionnaire asked about cell phone ownership, internet access, and media usage behaviors and requested contact information to send the incentive. Basic demographics, such as age, race/ethnicity, and household composition were collected. Respondents were able to complete the questionnaire by mail, phone, or on-line. The sample design included two waves and three different promised incentives, each of which increased in the second wave. Sample in the first wave was equally divided into three promised incentive groups: $5, $10, and $20. Wave two was initiated to non-respondents after 21 days. In this second wave, the promised incentives were increased to $10, $20, and $30, respectively. Mailings in both waves were identical with the exception of a $2 prepaid incentive included in wave one. The conditions of the promised incentive and amounts were made clear both in the cover letter and on the questionnaire itself. Results indicate which promised incentive amount produced the highest response in the fi rst wave, second wave, and combined, and whether statistical differences exist among the response rates. Percentages of returned surveys by mode – mail, phone, Web – will also be examined. Further, the demographic differences by incentive group, wave, and mode will be presented. The results of this study contribute substantial insight to understanding the effect of promised incentives on the response rates of mail surveys when phone contact cannot be made.

Prepays, Promises, and Postpays-Additional evidence on what helps response rates

Mathematica Policy Research

Low response rates continue to be a major issue for random digit dial (RDD) surveys. We recently tested a variety of incentive structures in a large nationwide RDD household survey, the Community Tracking Study (CTS). The three prior rounds of the CTS utilized a methodology that incorporated a panel of telephone numbers supplemented with new telephone numbers in each round. For the 2007 CTS, a new panel was begun, which meant that the benefit of cooperative households from prior rounds would not be present. To increase the overall response rate for the study, we experimented with both the timing and the quantity of respondent incentives. After randomly generating telephone numbers, we matched household addresses to those numbers where possible. The first experiment tested the effect of a cash prepay ($5 bill), in the initial mailing to those households with a matched address. Secondly, we experimented with the amount of a promised incentive- offering either $20 or $40 for completion of the survey. For households that refused over the telephone, we then sent checks to those with a matched name and address. Finally, we tested sending the $40 check as a prepay to some households. Preliminary results confirm previous studies as to the effectiveness of the $5 cash incentive in both increased response rates and reduced effort to gain completes. Final results of these and the other experiments will add to the literature on respondent incentives.

The Effect of Varying Incentives on Survey Results in an Age-Targeted Telephone Study

NORC

This paper presents the results of an experiment designed to examine the differential effect of varying incentive amounts on households with unlisted telephone numbers. The Neighborhood, Organization, Aging and Health project (NOAH) was a telephone study funded by the National Institute on Aging and conducted by NORC examining the influence of neighborhood context on the health and well-being of older adults. We completed approximately 1,500 interviews with individuals aged 65 and older living in selected Chicago communities. The project chose to purchase an age-targeted sampling frame from InfoUSA for efficiency. In order to optimize coverage, we also decided to include households believed to contain at least one 65-year-old but that were missing a telephone number. We then used a commercial locating product to retrieve telephone numbers for households with missing numbers. Locating was unable to retrieve telephones number for a majority of households missing telephone numbers a-priori. Finally, we sent a letter to those households we were unable to locate telephone numbers for informing them of the study and asking them to call in to participate. NORC designed an experiment to examine the effect of different incentives amounts ($10, $5, or No Incentive) on response rates for those households missing telephone numbers. This paper will discuss the results of the experiment. Specifically, we will address: whether or not respondents were willing to call-in to complete a telephone interview after being prompted by a letter; what incentive was most effective in encouraging respondents to call in; what type of households were willing to call in; and ultimately, was it viable to use a targeted commercial sample list including households without phone numbers to conduct a telephone study.