An Empirical Test of a Scale For Innovativeness

Citation:

C. Samuel Craig and James L. Ginter (1975) ,"An Empirical Test of a Scale For Innovativeness", in NA - Advances in Consumer Research Volume 02, eds. Mary Jane Schlinger, Ann Abor, MI : Association for Consumer Research, Pages: 555-562.

Advances in Consumer Research Volume 2, 1975      Pages 555-562

AN EMPIRICAL TEST OF A SCALE FOR INNOVATIVENESS

C. Samuel Craig, Cornell University

James L. Ginter, The Ohio State University

[This research was funded by the Division of Research, College of Administrative Science, The Ohio State University.]

[C. Samuel Craig is an Assistant Professor of Marketing at Cornell University.]

[James L. Ginter is Assistant Professor of Marketing, The Ohio State University.]

A scale developed to measure the trait of innovativeness was tested for predictive validity. The sample population of innovators were individuals who had purchased a new Mustang II early in the model year. Noninnovators were those individuals who purchased a new car other than Mustang II during the same time period. Results indicate that certain components of the scale (determined by factor analysis) did discriminate between sample populations.

INTRODUCTION

Research into the temporal (see Rogers, 1962 and Robertson, 1971) and spatial (see Brown, 1968) aspects of the diffusion of innovation has taken mary different tacks. Among the most intriguing studies, from the standpoint of consumer behavior, are those that attempt to identify characteristics of innovator populations versus non-innovator populations. Research delving into individual differences in the marketing literature has tended to focus on demographic characteristics, social interactions, and to a lesser extent, personality factors (Robertson, 1971, pp.100-101). Typically, personality factors account for a small and frequently non-significant portion of the variance in consumers' behavior. In a comprehensive review, KassarJian (1971) suggests that one of the major reasons for the lackluster performance of personality is the predilection for wholesale appropriation of pre-existing personality inventories that were not designed nor intended for marketing use.

The purpose of this study is to test the predictive validity of an innovativeness scale developed specifically to measure that trait. The actual development of the scale is covered elsewhere (Leavitt and Walton, 19743 and will not be dealt with here. The focus of this paper is to examine the usefulness of the scale in discriminating between a sample of innovators versus a sample of non-innovators.

THE DATA

Potential respondents' names were gathered from lists of people who had purchased 1974 model cars from September, 1973, to December, 1973. [Only individuals who had purchased a Mustang II, Pinto, Vega, Torino, or Ford were selected] During January, 1974, individuals were first called by telephone and asked if they were willing to participate in a study about automobiles. Those who were willing to participate were then sent a six-page questionnaire which contained the innovativeness scale along with a number of other questions. Data gathered on the questionnaire also included: automobile A-I-O's (activities, interest, and opinions), product use information, and demographic information. The questionnaire was mailed out to a total of 800 ne~ car buyers. Of the 420 returned questionnaires 324 had fully completed innovativeness scales. Only these were included in the analyses.

Rather than consider a multitude of consumer characteristics, analysis focused on the usefulness of the innovativeness scale in identifying innovators. Innovators were operationally defined as those individuals who had purchased a 1974 Mustang II from its introduction in September, 1973, until December, 1973. In terms of Robertson's (1971) continuum of innovation the Mustang II is considered to be dynamically continuous. It does not merely represent a model change but, rather, a departure from what the Mustang had evolved to since 1964. It is not, however, a radical departure from existing automobiles such as an electric or steam-powered car would represent.

An attempt was made to have both samples identical except for the indulgence in innovative behavior. The alternative approach of selecting 2 random sample of area residents to compare against innovators was rejected. It was felt that use of a random sample would create a situation where the alternative explanation, i.e., that new car buyers,in general, are different, would be tenable. By considering two samples, both of whom purchased new cars, any differences are more likely to be a function of unique characteristics of innovators.

RESULTS

The results section is divided into three parts. Initially, a factor analysis (using varimax rotation) of the innovativeness scale is presented to determine its components. Then, the raw scale values are examined first as summed innovativeness scores and then collectively as predictor variables in a multiple discriminant analysis. Finally, respondents' factor scores were used in place of the individual scale items as input into several multiple discriminate analyses.

Innovativeness Factors

The actual factors are of considerable interest as they provide some indication of the dimensions contained by this scale of innovativeness (see Table 1). Factor 1, "New is Wasteful," seems to reflect a general distrust of anything new and different. In a sense it is a tradition-oriented factor and is indicative of a general resistance to new things. Perhaps the most important aspect of this factor is the notion that new things are inherently wasteful.

Questions loading on Factor 2, "Social Desirability," all indicate an extreme tolerance for others. Individuals who would agree to all the questions would tend to be very acquiescent. These five questions were taken from Crowne and Marlowe (1 + 4) to detect those individuals who tend to give socially desirable responses to questions. As would be expected they all loaded together. Factor 3, "Novelty Seeking," indicates a general affinity for new things and new ideas. This factor would seem to be an important component of innovativeness, particularly when the product or idea is demonstrably new. Factor 4, "Risk Aversion," ought to be negatively associated with innovativeness. Individuals who are risk avoiders are not apt to be the first to try new products.

Factor 5, "Style Consciousness," is a component of innovativeness that seems to deal with people's awareness and concern about changing styles. Factor 6, "Satisfaction with Status Quo," on the other hand, suggests a desire to maintain the status quo. One puzzling aspect of this factor is the positive loading of, "When I see a new brand on the shelf, I often buy it just to see what it's like." It could indicate that there is a brand-specific component and also a component dealing with society (or less specific factors) in general. Finally, Factor 7, "Other-Directedness," deals with looking to others for advice about products. It also seems to reflect some uncertainty about the soundness of one's own ideas.

TABLE 1

FACTOR ANALYSIS OF INNOVATIVENESS SCALE  [The seven factors explained 52 percent of the variance. n = 353.]

Innovativeness Scale

The innovativeness scale was used to compare the respondents who had purchased a Mustang II with those who had purchased different cars during the same time period. The questions were rescaled so that all "more innovative" responses were in the same direction for al] questions. Responses to the questions were first summed and Mustang II owners' totals were compared with those of Pinto, Vega, Torino, and Ford owners. There was no significant difference between the two groups.

In order to utilize the information collected by the scale more fully, discriminant analysis was used to compare the two groups on the 25 questions. This analysis did not consider the structure of the scale identified through the previous factor analysis, but it did take into account the multivariate relationship of the responses to the 25 original questions. The results showed that 3 of the 25 items were significantly different for the two groups at the .05 level. Mustang II owners agreed less with the statements, "I am always willing to admit it when I make a mistake" and "I like to fool around with new ideas even if they turn out to be a waste of time." Note that these questions loaded most highly on factors 2 and 3. Mustang II owners agreed more with "I enjoy looking at new styles as soon as they come out." This question loaded most highly on factor 5.

Factor Components Discriminant Analyses

The rather weak performance of the summed scores and the 25 individual items in the preceding comparative anaLyses prompted the use of the seven previously identified factors. If the factors actually represent constructs within the scale, one would expect greater differences between the groups. Using the respondents' factor scores on the seven factors, a discriminant analysis was run comparing Mustang II purchasers with all the other new car purchasers. Three of the factors were significantly different for the two groups (see Table 2). The Mustang II owners were lower on factors 2 and 4 (Social Desirability and Risk Aversion) and higher on factor 6 (Satisfaction with Status Quo). This improved discriminatory performance of the scale seems to indicate that the constructs identified through the factor analysis were operative in the purchase of the newly introduced brand.

The hypothesis tested in this analysis was that the two groups are equal on the set of factor scores. Although the test of significance is the appropriate test of this hypothesis, it is also interesting to consider the predictive power. Table 3 shows that the group of Mustang II owners was predicted with 626 accuracy and the alternative group was predicted with 62% accuracy. It should be noted that these figures were developed on the set of data on which the discriminant function was developed rather than a hold-out sample.

TABLE 2

DISCRIMINANT ANALYSIS: MUSTANG II VERSUS PINTO, VEGA, TORINO, AND FORD USING FACTOR SCORES

TABLE 3

PREDICTION RESULTS: MUSTANG II VERSUS PINTO, VEGA, TORINO, AND FORD USING FACTOR SCORES

The possible confounding effect due to aggregation of purchasers of very different types of cars in the previous analysis was eliminated by comparing owners of Mustang II with those of Pinto and Vega, some of its primary competitors. The results as shown in Table 4 are very similar to those of the previous analysis.

TABLE 4

DISCRIMINANT ANALYSIS: MUSTANG II VERSUS PINTO AND VEGA USING FACTOR SCORES

A final comparison was conducted among the owners of brands manufactured by the Ford Motor Company. Mustang II owners were compared with those of the larger sized brands (Torino and Ford). The results (see Table 5) show that Mustang II owners were again lower on factor 2 (Social Desirability) and they were also higher on factor 5 (Style Consciousness).

TABLE 5

DISCRIMINANT ANALYSIS: MUSTANG II VERSUS TORINO AND FORD USING FACTOR SCORES

DISCUSSION

The summary of these results is interesting. In all of the comparative analyses, the Mustang II owners were lower on Social Desirability. The nature of these questions seems to indicate that these respondents tended to respond according to their feelings rather than responding in what might be deemed a more socially desirable way. They may also make decisions with less regard to what is expected of them and therefore accept more quickly (or seek out) a newly introduced brand. They also showed less risk aversion than owners of Pinto or Vega and Pinto, Vega, Torino, or Ford, This result is supportive in that those who were less averse to risk purchased the new brand. However, this factor was not significantly different between purchasers of Mustang II and Torino or Ford. This may be explained in part, since the purchasers of these three brands offered by the same company may have perceived little inter-brand risk. An additional factor which was significantly different between these two groups was Style Consciousness, with the Mustang II owners scoring higher. This result is consistent with Ford's marketing objectives in offering a sporty, luxurious, and stylish economy car.

This application of the scale is also interesting in that in this instance, the factors associated with the wastefulness of new things, novelty seeking, and other-directedness were not significantly different for purchasers of the Mustang II. The Mustang II may not have been considered sufficiently new and different to be perceived as wasteful. Further, it may not have been perceived as sufficiently novel to bring out the novelty-seeking component. Also, in light of the earlier version of the Mustang, people may not have felt the judgment of others was essential (or non-essential) for the decision process. Given the magnitude of a new car purchase, an alternative explanation may be that other factors outweighed those related to the newness of the brand in the decision Process.

CONCLUSION

Certain components of the innovativeness scale appear to discriminate between innovators and non-innovators. In another situation with a different type of innovation, the other components might be significant. The results of the analysis suggest three conclusions. First, the innovativeness scale constructed by Leavitt and Walton has some predictive validity. The fact that a substantial proportion of the constructs identified within the scale were significantly different for Mustang II owners lends support for the scale and warrants further research on its application to other purchase situations. Second, the trait of innovativeness is not a homogeneous construct, but a conglomerate of a number of constructs. Further research, using different product categories, ought to be directed at determining which components of innovativeness are operating in different situations. Once this is established, marketers will have a firmer basis for identifying potential innovators and designing marketing strategies aimed at facilitating adoption of new products. Finally, the significant differences obtained strongly suggest the utility of developing scales aimed at measuring phenomena of interest to marketers. This would appear a far more fecund approach than attempting to apply less suitable extant scales.

REFERENCES

Brown, L. A. Diffusion dynamics. Lund, Sweden: Lund Studies in Geography, 1968.

Crowne, D. P. & Marlowe, D. The approval motive: Studies in evaluative dependence. New York: Wiley, 1964.

Kassarjian, H. H. Personality and consumer behavior: A review. Journal of Marketing Research, 1971, 8, 409-418.

Leavitt, C. L. & Walton, J. Development of a scale for innovativeness. Working paper, The Ohio State University, 1974.

Robertson, T. S. Innovative behavior and communication New York- Holt, Rinehart, and Winston, 1971.

Rogers, E. M. & Shoemaker, F. F. Communication of innovations. New York: The Free Press, 1971.

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