The Effects of Salutation, Monetary Incentive, and Degree of Urbanization on Mail Questionnaire Response Rate, Speed, and Quality

Linda L. Golden, University of Texas at Austin
W. Thomas Anderson, University of Texas at Austin
Louis K. Sharpe, Sharpe's Department Store, Checotah, Oklahoma
ABSTRACT - The effects of three cover letter personalization treatments, a cash drawing monetary incentive, and rural, suburban, and urban place of residence upon response rate, speed, and quality for a mail questionnaire were investigated. The results indicated that cover letter personalization treatments did not have significantly different effects on response rate, speed, or quality. The probabilistic monetary incentive lowered response rate, but degree of urbanization moderated the effects on response rate, speed, and quality. Rural, suburban, and urban respondents may be differentially misunderstood by mail survey researchers due to potentially differing degrees of response and nonresponse biases.
[ to cite ]:
Linda L. Golden, W. Thomas Anderson, and Louis K. Sharpe (1981) ,"The Effects of Salutation, Monetary Incentive, and Degree of Urbanization on Mail Questionnaire Response Rate, Speed, and Quality", in NA - Advances in Consumer Research Volume 08, eds. Kent B. Monroe, Ann Abor, MI : Association for Consumer Research, Pages: 292-298.

Advances in Consumer Research Volume 8, 1981      Pages 292-298

THE EFFECTS OF SALUTATION, MONETARY INCENTIVE, AND DEGREE OF URBANIZATION ON MAIL QUESTIONNAIRE RESPONSE RATE, SPEED, AND QUALITY

Linda L. Golden, University of Texas at Austin

W. Thomas Anderson, University of Texas at Austin

Louis K. Sharpe, Sharpe's Department Store, Checotah, Oklahoma

ABSTRACT -

The effects of three cover letter personalization treatments, a cash drawing monetary incentive, and rural, suburban, and urban place of residence upon response rate, speed, and quality for a mail questionnaire were investigated. The results indicated that cover letter personalization treatments did not have significantly different effects on response rate, speed, or quality. The probabilistic monetary incentive lowered response rate, but degree of urbanization moderated the effects on response rate, speed, and quality. Rural, suburban, and urban respondents may be differentially misunderstood by mail survey researchers due to potentially differing degrees of response and nonresponse biases.

INTRODUCTION

One of the most commonly used and intensively researched data gathering techniques in the behavioral sciences is the mail questionnaire. However, as two of the most recent review articles point out (Kanuk and Berenson 1975, p. 451; Linsky 1975, p. 100), even with the large volume of research on mail questionnaires, very few systematic conclusions can be drawn about the effectiveness of the various mail survey techniques. This is partially due to inconsistent results across studies, a lack of replication and research on any one technique, a near absence of research designed specifically to compare mail survey techniques in a variety of purposively selected populations, and the lack of a systematic body of knowledge or conceptual framework developed from general theory (Kanuk and Berenson 1975, Linsky 1975).

While the present research does not constitute a replication, nor do the authors pretend to develop a theoretical framework for mail survey research, this investigation does provide additional information on the effectiveness of two mail survey techniques: the personalized cover letter and the cash drawing as a monetary incentive. Moreover, this research compares the effects for respondents from three purposively selected populations: urban, suburban, and rural.

Personalization of Cover Letters

While one review article (Linsky 1975, p. 92) reports 16 studies dealing with personalization of cover letters, a review of these articles and others reveals that the method of personalization varies widely across studies. Personalization treatments have ranged from individually typed letters versus form letters (Simon 1967) to handwritten postscripts urging reply versus no postscript (Frazier and Bird 1958). Similar to figures reported by Linsky (1975, p. 92), of the 20 studies reviewed, 11 reported higher response rates for personalized over impersonal cover letters (Carpenter 1974, Cox, Anderson, and Fulcher 1974, Frazier and Bird 1958, Linsky 1965, Longworth 1953, Moore 1941, Roeher 1963, Simon 1967), five reported higher response rates for impersonal cover letters (Andreason 1970, Houston and Jefferson 1975, Simon 1967, Watson 1965, Weilbacher and Walsh 1952), and four studies reported no significant differences between personalized and impersonal cover letters (Clausen and Ford 1947, Kawash and Aleamoni 1971, Kimball 1961).

Thus, results for the various forms of cover letter personalization are mixed. Further, the results are also mixed for any particular form of personalization, although almost half as many different forms of personalization were tested as there are number of studies.

The present research focuses on personalization via the inside address and cover letter salutation. Five studies were found which investigated the effects of personalization via either the salutation or the inside address and salutation on response rate. One reported significant differences between personalized and impersonal cover letters favoring the personalized form (Cox, Anderson, and Fulcher 1974), one reported significant differences favoring the impersonal form (Andreason 1970), and three reported no significant differences between personalized and impersonal cover letters (Clausen and Ford 1947, Kimball 1961, and Weilbacher and Walsh 1952). While the results across studies generally indicate no significant effect of personalization, the inconsistent results imply that the research context and topic may be an important factor in the effects of personalization.

Monetary Incentives

Inducements offered to the respondent in an attempt to increase response rate have frequently been in the form of monetary incentives. While studies investigating the effectiveness of premiums (nonmonetary rewards such as a ballpoint pen or trading stamps) have unanimously reported that offering a premium produces a higher response rate than offering no premium at all (Linsky 1975, p. 99), one study found that inclusion of money significantly increased response rate over inclusion of a premium (Goodstadt, Chung, Kronitz, and Cook 1977). "In general, money seems to be the most effective and least biasing incentive, the easiest to obtain and mail, and the most useful to all recipients" (Kanuk and Berenson 1975, p. 447).

There are two broad categories of studies investigating response rates produced by monetary rewards: those investigating a prepaid monetary incentive and those investigating a promised monetary incentive. Of the 20 studies reviewed researching the effects of prepaid monetary incentives, all found that offering a prepaid incentive produced a higher response rate than offering no monetary incentive at all (Armstrong and Overton 1971, Bevis 1948, Blumberg, Buller, and Hare 1974, Erdos 1970, Goodstadt, Chung, Kronitz, and Cook 1977, Hackler and Bourgette 1973, Buck and Gleason 1974, Hancock 1940, Kephart and Bressler 1958, Kimball 1961, Maloney 1954, Newman 1962, Pressley and Tullar 1977, Shuttleworth 1931, Watson 1965, Wiseman 1973, Wotruba 1966). Not all of these studies employed significance tests or reported significant differences; however, it may be concluded unequivocally that a prepaid monetary incentive produces a higher percentage response than offering no monetary incentive. And, as one review article reports, the larger the prepaid monetary incentive, the greater the increase in response rate (Armstrong 1973, p. 111).

Of the five studies investigating the effects of a promised monetary incentive on response rate, three compared promised, prepaid, and no incentive treatments (Blumberg, Fuller, and Hare 1974; Hancock 1940; Wotruba 1966). All three found small and insignificant differences between promised incentives and a no incentive control, but, as reported above, much larger differences between prepaid incentives and the control. In addition, while promised incentives increased the response rate only very slightly, the prepaid treatment produced a significantly higher response rate than did the promised incentive treatment. One study by Gelb (1975), which did not include a control group, found the prepaid monetary incentives significantly increased response rate over promised monetary rewards for middle-class subjects while promised rewards produced a higher response rate than prepaid incentives for lower-class subjects. Schewe and Cournoyer (1976) found that against a control of no monetary incentive, the size of the promised monetary reward influenced its effectiveness. Thus, while promised incentives appear to be generally less effective than prepaid monetary rewards, the amount of the promised incentive (Schewe and Cournoyer 1976) and the population sampled (Gelb 1975) may alter or even reverse the general ineffectiveness of the promised monetary incentive.

Research Focus and Purpose

While response rate differences have been examined for a variety of populations, no studies were found in the literature comparing the possible differential responsiveness of rural, suburban, and urban residents to personalized cover letters and monetary incentives. Moreover, no study was identified which compared rural, suburban, and urban mail survey response rates, or any other measure of mail questionnaire response for these populations. These residential groups would be expected to have distinct socioeconomic and demographic profiles; such respondent differences have been shown to correlate with measures of response and nonresponse bias (see Kanuk and Berenson 1975).

The general purpose of this study is to investigate the effects of a cash drawing as a monetary incentive, cover letter personalization and degree of urbanization (area of residence) upon response rate, quality, and speed. Response speed and quality, or degree of questionnaire completion, not only provide additional insight into the response patterns related to the treatments and residential conditions but are also frequently-used measures of response bias. The cash drawing incentive and cover letter personalization treatments are expected to produce higher response rates, speed, and quality than no incentives or no personalization. And, it is expected that urban and suburban residents will have higher response rates, speed, and quality than rural residents. The research also investigates psychographic differences among respondents receiving the monetary incentive who entered and who did not enter the cash drawing, and psychographic differences associated with response speed and quality. It is expected that external locus of control and future-oriented respondents will be more likely to participate in the cash drawing and will have higher response, speed, and quality.

METHOD

The questionnaire was a three-page inquiry into shopping patterns for basic types of products: groceries, clothing, and furniture and appliances. Since the respondents were likely to be of varied educational backgrounds, the questionnaire was extensively pretested in the field for clarity.

Standard demographic information was obtained and four psychographic measures were included in the final sections of the questionnaire. The psychographic variables measured were: James' Internal-External Locus of Control (Rotter 1966), Life Satisfaction (Robinson 1977), Attitude Toward the Past, Present and Future (Rokeach 1956), and Traditional Family Ideology (Levenson and Huffman 1955). Life satisfaction and traditional family ideology measures were included only for descriptive comparison among rural, suburban, and urban residents; however, it was expected that locus of control and attitude toward past, present, and future would influence responsiveness to the monetary incentive offered.

The questionnaire was mailed to a sample of 4,500 potential respondents in Eastern Oklahoma, randomly selected from current telephone directories of five trading areas: (Tulsa (n = 1,500), suburbs of Tulsa (n = 1,500), Muskogee (n = 500), Chocotah (n = 500), and Stigler (n = 500). The last three communities represent the most rural communities sampled. The cover letter, each individually hand-signed, requested that the questionnaire be completed by the person who does most of the shopping in the household.

The two experimental treatments tested in this study, monetary incentive and salutation, were administered via the cover letter. Three thousand of the 4,500 potential respondents received a cover letter which described a drawing for cash prizes totaling $75.00. One $25.00 prize, five $5.00 prizes, and 25 $1.00 prizes were offered. The respondent was to return a cash drawing card in the postage-paid return envelope provided for the questionnaire if s/he wished to enter the drawing. The cash drawing card was used as the method for entry to allow the respondent to return the questionnaire anonymously without entering the contest. The dollar values of the prizes offered were chosen in order to maximize the respondents' probability of winning a potentially higher dollar-value return. Fifteen hundred potential respondents received a cover letter that made no mention of a monetary incentive but which, in all other aspects, paralleled the monetary incentive cover letter.

The cover letters also contained one of three different levels of salutation. Fifteen hundred potential respondents received a standardized "Dear Shopper" salutation with no inside address. Another 1,500 potential respondents were sent a cover letter with a formally typed complete inside address and "Dear Mr. and Ms.____" salutation including the respondent's last name. The remaining 1,500 respondents received a handwritten "Dear Mr. or Ms.____" salutation personalized with the respondent's last name, but with no inside address. The salutation treatments were designed to represent impersonal, formal, and personal salutations, respectively, although the envelopes were all typed and personalized with Mr. or Ms. . Respondents were randomly assigned to both the monetary incentive and salutation treatments; assignment was controlled so that each area of residence received the same proportion of each treatment.

Responses were evaluated on three criteria: response rate, speed, and quality. As the measure of success typically used by researchers, response rate was the percentage of the total mailing which was returned, regardless of degree of questionnaire completion. Response speed was measured continuously by the postmark on the respondents' return envelope; the earliest postmark was assigned a value of one. Questionnaires were mailed simultaneously from central Texas, and returns began within a week of the mailing; the range for the response speed variables was from one to 43 days. Response quality was evaluated according to the degree of completion of the questionnaire. Since the questionnaire contained four sections, response quality ranged from one to five: one being a fully-completed questionnaire, four representing only one part completed, and five a questionnaire with only scattered responses and no sections complete.

ANALYSIS AND RESULTS

A total of 833 questionnaires were returned, resulting in a response rate of 18.5 percent. Due to financial constraints, and in order to reach a broader segment of the population, the researchers opted to send questionnaires to a larger sample frame rather than reduce the sample frame size and use a follow-up. The response rate is reflective of what would be expected under no follow-up conditions.

Response Rate

Chi squares were computed for response and nonresponse frequencies for monetary incentive, salutation, and urbanization treatments separately and are presented in Table 1. Surprisingly, the nonmonetary level (control) produced a significantly higher response rate than did the monetary " SIZE="5incentive. Slightly over 20 percent of the 1,500 potential respondents who were mailed the nonmonetary cover letter returned the survey, whereas 17.7 percent of the 3,000 potential respondents mailed the monetary incentive returned the survey. Thus, the cash drawing was significantly less effective than offering no monetary incentive.

TABLE 1

EFFECTS OF INCENTIVE, SALUTATION, AND URBANIZATION ON RESPONSE RATE

There was no significant effect for the salutation treatment; all three levels yielded response rates approximating the expected 18.5 percent. However, response rate did vary significantly by urbanization. Urban residents had the highest response rate (21.3 percent) and were the only group whose frequency of response was higher than the expected rate (16.5 percent), while the rural residents had the lowest response rate (15.2 percent).

To investigate psychographic differences among participants and nonparticipants in the cash drawing, the data were submitted to one-way analysis of variance (Nie et al 1975, p. 422) for each psychographic variable separately. As indicated in Table 2 and as expected, locus of control and attitude toward past, present, and future were the only two psychographic variables on which participants and nonparticipants in the cash drawing differed. Respondents returning the cash drawing card were significantly more externally-oriented than respondents who did not return the cash drawing card. In addition, participants in the cash drawing were significantly more future-oriented than nonparticipants.

To investigate differential responsiveness to the personalization and monetary treatments among rural, suburban, and urban residents, chi squares were computed for response and nonresponse frequencies on each treatment separately. Of the six chi squares computed, only one was significant within an alpha level of five percent. The suburban response rate to the monetary incentive was 12.6 percent, compared to a 16.6 percent response rate for the control group (chi square of 4.22 with one degree of freedom). Thus, suburbanites responded less frequently than expected (13.9 percent) to the monetary incentive and more frequently than expected (14.0 percent) when there was no monetary incentive. While the suburban response rate was significantly decreased by the monetary incentive, rural and urban response rates did not differ between the monetary incentive and the control. Urbanization groups did not respond significantly differently to the cover letter personalization treatment levels.

Response Speed and Quality

In order to investigate the effects of the monetary incentive and salutation treatments and the urbanization condition upon response speed, the data were submitted to three-way analysis of variance (Nie et al 1975, p. 410). Table 3 shows the results of these analyses.

Neither treatment produced a significant main effect, and there were no significant interaction effects. There was, however, a significant main effect of urbanization. Contrary to expectations, rural respondents returned the questionnaire more quickly than either the suburban or the urban respondents, respectively. Scheffe tests (Scheffe 1959, p. 477) indicated that both rural and suburban respondents responded significantly more quickly than did urban respondents, although there was no significant difference in mean response speed between rural and suburban respondents. Thus, while urban respondents had the highest response rate, they were the slowest to respond. Rural respondents showed just the reverse pattern: the lowest response rate but the quickest response speed.

TABLE 2

PSYCHOGRAPHICS FOR RESPONDENTS RETURNING AND NOT RETURNING THE CASH DRAWING CARD

TABLE 3

EFFECTS OF INCENTIVE, SALUTATION, AND URBANIZATION ON RESPONSE SPEED

TABLE 4

EFFECTS OF INCENTIVE. SALUTATION, AND URBANIZATION ON RESPONSE QUALITY

TABLE 5

RELATIONSHIPS BETWEEN PSYCHOGRAPHICS AND RESPONSE SPEED AND QUALITY

Table 4 presents the results of the three-way analysis of variance (Nie et al 1975, p. 410) for the effects of the monetary incentive and salutation treatments and the urbanization condition on response quality. Again, there were no significant main effects for monetary incentive or salutation treatments and no significant two- or three-way interactions. Urbanization was the only variable which produced a significant main effect. The urban respondents had the highest response quality, followed by the suburban and rural respondents, respectively. Scheffe tests indicated that the rural respondents had a significantly lower quality of response than both the suburban and the urban respondents. There was no significant difference between urban and suburban response quality. Thus, while urban respondents were relatively slower respondents, they were relatively thorough in their questionnaire completion. Rural respondents, on the other hand, were relatively more quick to respond, but questionnaire completion was lower. The suburban respondents continually fell in the middle, aligning themselves with the rural respondents in terms or speed and the urban respondents in terms of quality. Suburban response rate also tended toward the lower response rate of the rural residents.

The data for the psychographic variables and response speed and quality were submitted to correlational analyses (Nie et al 1975, p. 280). Table 5 presents the results of these analyses. Response speed was significantly correlated with locus of control and family ideology, although the Pearson correlations were not high. Response speed was significantly correlated with three of the four psychographic variables: locus of control, family ideology, and attitude toward the past, present, and future. Life satisfaction was not significantly related to either response speed or quality. The correlations of the significant psychographics with response quality were much higher than the correlations with response speed.

The more external the respondent's locus of control, the slower to respond and the lower the response quality. Respondents with a more internal locus of control responded more quickly and had a generally higher degree of questionnaire completion than did the respondents who were relatively more external. Further, the more traditional the family ideology, the faster the response speed and the lower the quality of response. The respondents with the least traditional family ideologies responded more slowly but with a higher completion rate than did the respondents with a relatively more traditional family ideology. For response quality, the. more past-oriented the respondent, the lower the quality. Thus, the more future-oriented respondents provided the most complete questionnaires; however, this psychographic was not significantly associated with response speed.

CONCLUSIONS AND IMPLICATIONS

This study investigated the effects of cover letter personalization, a promised and uncertain monetary incentive, and degree of urbanization upon response rate, speed, and quality for a mail questionnaire. The results suggest that the additional costs of cover letter personalization are not warranted if the researcher's objective is to increase response rate, speed, or quality. While the conclusions drawn from this study must be restricted to the sample under investigation and the specific treatments used, these results are consistent with those obtained by three of the five studies using the salutation as the method of personalization (Clausen and Ford 1947, Kimball 1961, Weilbacher and Walsh 1952). In addition, this study indicates that rural, suburban, or urban residence does not influence the effects of personalization.

A cash drawing monetary incentive appears to decrease response rate in contrast with no monetary incentive. There may be a raffle effect which results from this type of monetary, incentive, whereby potential respondents may regard the questionnaire as just another piece of "junk mail," decreasing the researcher's credibility. However, there were no differences between the monetary treatment and the control for response speed and quality. Thus, according to this study, the uncertain promised monetary incentive has no redeeming virtues.

Rural and urban respondents do not appear to be differentially responsive to uncertain monetary incentive treatments, but suburban respondents may be, as they returned the control treatment questionnaires at a significantly higher rate than the monetary incentive questionnaires. Thus, while rural and urban respondents did not tend to react negatively to the cash drawing, at least as evidenced by their response rate, suburban respondents evidenced the strongest "raffle effect reaction."

Respondents who received a "raffle" monetary incentive and returned the cash drawing card appear to be different from those who do not return the cash drawing card. Respondents returning the cash drawing card were significantly more external in their locus of control and significantly more future-oriented than nonparticipants in the cash drawing. Externally- and future-oriented persons may be more "raffle prone" and more likely to take a "gamble" than persons with other orientations, thus making them the most likely to be responsive to the type of incentive investigated here. If the researcher has prior knowledge of a sample's psychographic characteristics, an uncertain monetary incentive may be able to reduce the costs of a monetary "reward" and possibly increase response rate.

Urbanization appears to have an important and significant influence upon response rate, speed, and quality. Urban residents had the highest response rate, were slowest to respond but provided the highest quality responses. Rural residents were a mirror image of the urban residents, providing the lowest response rate, the quickest response, and the lowest response quality. Consistent with their mid-range urbanism, suburban respondents occupied the middle. Suburban respondents aligned themselves with rural respondents on response speed and with urban respondents on response quality.

Response rate and quality differences among rural, urban, and suburban residents may be partially explained by demographic differences, specifically education. [Although not formally reported in this paper, Chi Square analysis for 11 standard demographics resulted in significance at the .05 level for nine demographics. For educational level of chief wage earner (Chi Square = 75.07), urbanites were the most formally educated, ruralites the least, and suburbanites in between.] With the highest educational level, urban residents may be more experienced communicators resulting in a higher response rate and response quality than for either suburbanites or ruralites. In addition, the ruralites, with the lowest educational level, would be expected to be the least responsive to mail surveys both in terms of response rate and quality. Such was the case.

Response speed and quality provide some response bias information for the three residential groups. The late respondents to a questionnaire sampling the three urbanism groups were more likely to be urban; thus, the nonrespondents may tend to be more like the urban population sampled although the urban residents do respond at a higher rate. Because of their low quality of response, however, the researcher is likely to obtain the least information from rural samples. Coupled with their low response rate, the low quality of responses make these respondents the least understood, or most misunderstood, of the three groups. They are, in fact, the least represented in this study both in terms of response and nonresponse bias.

Response speed and quality tend to be associated with certain psychographic characteristics. Respondents who were the slowest to respond were more external in their locus of control and less traditional in their family ideology. In addition, respondents who provided the lowest response quality in terms of survey completion were more externally-oriented and more traditional in family ideology, as well as being more past-oriented than respondents with higher quality responses. Thus, without taking place of residence into consideration, the nonrespondents (to the extent that they are similar to late respondents) are likely to be more external in their locus of control and less traditional than respondents. However, the measure of response bias indicates that the least complete questionnaires are likely to come from more externally-oriented respondents and those with a more traditional family ideology.

The resounding implication of this study is the importance of the population sampled and the subsequent effect upon mail questionnaire response rate and potential bias in terms of response speed and quality. Researchers sampling across two or more of these populations are advised that they may have differing response patterns which should be taken into consideration in adjusting for any potential biases. Rural populations may be the most misunderstood, while urban populations may be the most represented and the least misunderstood. Wave mailing of questionnaires should also consider the unique characteristics of these populations when determining the timing of questionnaire waves. Rural, suburban, and urban samples cannot be considered to be drawn from the same populations. Although potentially the least costly, mail questionnaires may be the least effective means of sampling rural populations.

REFERENCES

Andreasen, Alan R. (1970), "Personalizing Mail Questionnaire Correspondence," Public Opinion Quarterly, 34, 273-7.

Armstrong, J. Scott (1975), "Monetary Incentives in Mail Surveys," Public Opinion Quarterly, 39, 111-16.

Armstrong, J. Scott, and Overton, Terry S.(1971), "Brief vs. Comprehensive Descriptions in Measuring Intentions to Purchase," Journal of Marketing Research, 8, 114-17.

Bevis, Joseph C. (1948), "Economic Incentives Used for Mail Questionnaires," Public Opinion Quarterly, 12, 492-3.

Blumberg, Herbert H., Fuller, Carolyn and Hare, Paul (1974), "Response Rates in Postal Surveys," Public Opinion Quarterly, 38, 113-23.

Carpenter, E. W. (1974), "Personalization in Mail Surveys of the General Public: A Reassessment in Light of Recent Innovations," Public Opinion Quarterly, 38, 116-20.

Clausen, John A., and Ford, Robert N.(1947), "Controlling Bias in Mail Questionnaires," Journal of the American Statistical Association, 42, 497-511.

Cox, Eli P. III, Anderson, Jr., W. Thomas and Fulcher, David G.(1974), "Reappraising Mail Survey Response Rates," Journal of Marketing Research, 11, 413-7.

Erdos, Paul L. (1970). Professional Mail Surveys, New York: McGraw-Hill.

Frazier, George, and Bird, Kermit (1958), "Increasing the Response of a Mail Questionnaire," Journal of Marketing, 22, 186-7.

Gelb, Betsy D. (1975), "Incentives to Increase Survey Returns: Social Class Considerations," Journal of Marketing Research, 11, 107-9.

Goodstadt, Michael S., Chung, Linda, Kronitz, Reena and Cook, Gaynoll (1977), "Mail Survey Response Rates: Their Manipulation and Impact," Journal of Marketing Research, 14, 391-5.

Hackler, James C., and Bourgeois, Patricia (1973), "Dollars, Dissonance, and Survey Returns," Public Opinion Quarterly, 37, 266-81.

Hancock, John (1940), "An Experimental Study of Four Methods of Measuring Unit Costs on Obtaining Attitude toward the Retail Store," Journal of Applied Psychology, 24, 213-30.

Houston, Michael, and Jefferson, Robert (1975), "The Negative Effects of Personalization on Response Patterns in Mail Survey," Journal of Marketing Research, 7, 114-7.

Huck, Schuyler W., and Gleason, Edwin M. (1974), "Using Monetary Inducements to Increase Response Rates from Mail Surveys," Journal of Applied Psychology, 59, 222-5.

Kanuk, Leslie, and Berenson, Conrad (1975), "Mail Surveys and Response Rates: A Literature Review," Journal of Marketing Research, 12- 440-53.

Kawash, Mary B., and Aleamoni, Lawrence M. (1971), "Effect of Personal Signature on the Initial Rate of Return of a Mailed Questionnaire," Journal of Applied Psychology, 55, 569-92.

Kephart, William M., and Bressler, Marvin (1958) "Increasing the Response to Mail Questionnaires: A Research Study," Public Opinion Quarterly, 22, 123-32.

Kimball, Andrew E. (1961), "Increasing the Rate of Return in Mail Surveys," Journal of Marketing, 25. 63-5.

Levenson, Daniel, and Huffman, Phyllis(1955), "Traditional Family Ideology and Its Relation to Personality," Journal of Personality, 24, 251-73.

Linsky, Arnold S. (1965), "A Factorial Experiment in Inducing Responses to a Mail Questionnaire," Sociology and Social Research, 49, 183-9.

Linsky, Arnold S. (1975), "Stimulating Responses to Mail Questionnaires: A Review," Public Opinion Quarterly, 39, 82-101.

Longworth, Donald S. (1953), "Use of a Mailed Questionnaire, "American Sociology Review, 18, 310-13.

Maloney, Paul W. (1954), "Comparability of Personal Attitude Scale Administration With Mail Administration With and Without Incentive," Journal of Applied Psychology, 38, 238-9.

Newman, Sheldon W. (1962), "Differences Between Early and Late Respondents to a Mail Survey," Journal of Advertising Research, 2, 27-39.

Nie, Norman H., Hull, C. Hadlai, Jenkins. Jean G., Steinbrenner, Karen and Bent, Dale H. (1975). Statistical Package for the Social Sciences, Second Edition, New York: McGraw-Hill.

Pressley, Milton M., and Fuller, William L. (1977), "A Factor Interactive Investigation of Mail Survey Response Rates from a Commercial Population," Journal of Marketing Research, 14, 108-11.

Robinson, John P. (1977). How Americans Use Time: A Social-Psychological Analysis of Everyday Behavior, New York: Praeger Press.

Roeher, Allan G. (1963), "Effective Techniques in Increasing Response to Mailed Questionnaires," Public Opinion Quarterly, 27, 299-302.

Rokeach, Milton (1956), "Political and Religious Dogmatism: An Alternative to the Authoritarian Personality," Psychological Monographs, 70, No. 425.

Rotter, 3. B. (1966), "Generalized Expectancies for Internal Versus External Control of Reinforcement," Psychological Monographs, 80, No. 609.

Scheffe, Henry (1959). Analysis of Variance, New York: John Wiley and Sons.

Schewe, Charles D., and Cournoyer, Norman G.(1976), "Prepaid vs. Promised Monetary Incentives to Questionnaire Response: Further Evidence," Public Opinion Quarterly, 40, 105-7.

Shuttleworth, Frank K. (1931), "A Study of Questionnaire Technique," Journal of Educational Psychology, 22, 652-8.

Simon, Raymond (1967), "Response to Personal and Form Letters in Mail Surveys," Journal of Advertising Research, 7, 28-30.

Watson, John J. (1965), "Improving the Response Rate in Mail Research," Journal of Advertising Research, 5, 48-50.

Weilbacher, William, and Walsh, H. Robert (1952), "Mail Questionnaires and the Personalized Letter of Transmittal," Journal of Marketing, 16, 331-6.

Wiseman, Frederick (1976), "A Reassessment of the Effects of Personalization on Response Patterns in Mail Surveys," Journal of Marketing Research, 13, 110-1.

Wotruba, Thomas E. (1966), "Monetary Inducements and Mail Questionnaire Response," Journal of Marketing Research, 3, 398-400.

----------------------------------------