A Study of Individual Factors Explaining Movie Goers’ Consultation of Film Critics

ABSTRACT - This research focuses on some key individual factors explaining the extent of movie goers’ consultation of film critics. Four research hypotheses are put forward on the basis of previous studies in the area of consumer information search and three of them are supported by means of a survey among 120 amateurs of cinema. The results indicate that the consultation of film critics is positively associated with movie goers’ susceptibility to social influence, negatively associated with their self-esteem and negatively associated with their personal involvement with cinema. In addition, the consultation of film critics is higher among those who declare being more knowledgeable about cinema, although consultation decreases at the highest levels of subjective knowledge. Most of the results agree with findings reported in previous research on consuer information search.

Citation:

Alain d’Astous (1999) ,"A Study of Individual Factors Explaining Movie Goers’ Consultation of Film Critics", in E - European Advances in Consumer Research Volume 4, eds. Bernard Dubois, Tina M. Lowrey, and L. J. Shrum, Marc Vanhuele, Provo, UT : Association for Consumer Research, Pages: 201-207.

European Advances in Consumer Research Volume 4, 1999      Pages 201-207

A STUDY OF INDIVIDUAL FACTORS EXPLAINING MOVIE GOERS’ CONSULTATION OF FILM CRITICS

Alain d’Astous, Ecole des HEC, Canada

[The author thanks Nadia Touil for her help in collecting the data and Pierre Balloffet for many useful comments made on this paper.]

ABSTRACT -

This research focuses on some key individual factors explaining the extent of movie goers’ consultation of film critics. Four research hypotheses are put forward on the basis of previous studies in the area of consumer information search and three of them are supported by means of a survey among 120 amateurs of cinema. The results indicate that the consultation of film critics is positively associated with movie goers’ susceptibility to social influence, negatively associated with their self-esteem and negatively associated with their personal involvement with cinema. In addition, the consultation of film critics is higher among those who declare being more knowledgeable about cinema, although consultation decreases at the highest levels of subjective knowledge. Most of the results agree with findings reported in previous research on consuer information search.

INTRODUCTION

Consumers commonly gather information from expert sources in order to form evaluations and make better decisions. When information is perceived as insufficient in a given consumption situation, the consultation of knowledgeable sources is usually an efficient way of reducing uncertainty (Hugstad, Taylor and Bruce 1987). In this article, we are interested in some key individual factors explaining movie goers’ consultation of expert sources. Movies are highly visible and very popular experiential products that are reviewed, analyzed and critiqued by numerous people. Although they are not perceived as the most influential information sources by movie goers (Faber and O’ Guinn 1984), film critics do play a significant role in the launching of new movies (Farber 1976). Identifying the individual characteristics that distinguish those amateurs of cinema who consult film critics from those who do not is a relevant research goal.

RESEARCH ON THE INFLUENCE OF FILM CRITICS

Because they are among the first to see new movies and because their role is to share with the public their movie consumption experiences, film critics can be considered both as innovators and opinion leaders (Reddy, Swaminathan and Motley 1996). But, what influence do they have on movie goers’ a priori evaluations and intentions to see, or to not see, a new movie ? There is surprisingly few studies that have addressed this question.

Wyatt and Badger (1984) presented positive, negative and equivocal movie reviews to a sample of university students and found that expressed interest towards a movie was generally consistent with the opinion found in its review. Thus, interest in seeing a movie was higher when the review was positive than when it was negative. It must be noted that in this research, the movie review was the only information available to subjects.

Boor (1992) found positive correlations between the ratings given to 568 movies by six reputable film critics and movie goers’ evaluations as compiled by Consumer Reports. The strength of the correlations was such that the author concluded that in general film reviews are good indicators of the value of movies. Although Boor’s (1992) results are interesting, it must be kept in mind that they are correlational and aggregated.

A recent research by Eliashberg and Shugan (1997) has examined the predictive power of film reviews with regard to box office sales. These researchers found that film reviews were correlated with late and cumulative box office sales, but not with sales observed during the first four weeks. They concluded that although film reviews can predict the success of a new movie, they do not necessarily have an influence on movie goers. In other words, the observed correlation between film reviews and sales would primarily reflect the intrinsic value of movies.

VTzina (1997) used an experimental approach to study the impact of critics’ quotes in movie advertising (e.g., "Terrific! The best movie so far this year !") on intentions to see a new movie. He found that the number of quotes (0, 2, 4 or 6) did not impact significantly on movie goers’ intentions. In his study, movie genre (action, drama, etc.) was the best predictor of people’s intention to see a new movie.

The empirical evidence concerning the influence of film critics on movie goers’ evaluations and intentions is not compelling. Upon examination of the few published studies, one must conclude that researchers have often used correlational and aggregated methods (Boor 1992; Eliashberg and Shugan 1997). When experimental methods at an individual level wer used, researchers have neglected to present any other information than the movie review itself (Wyatt and Badger 1984) or have been simply unable to obtain significant effects of critics’ opinions (VTzina 1997).

Although research on film critics’ influence on movie goers’ evaluations and intentions is certainly relevant, it would seem interesting before looking at the impact of the information diffused by film critics to identify the variables that can predict the extent of consultation of film critics by movie goers. Using a hierarchy of effects perspective (Lavidge and Steiner 1961), the impact of information on consumers necessarily follows from the information acquisition phase. In the next section, we present a set of individual characteristics aimed at predicting movie goers’ consultation of film critics.

DEVELOPMENT OF RESEARCH HYPOTHESES

The consultation of film critics can be seen as a special case of consumer information acquisition. It is therefore pertinent to examine the results of studies conducted on this topic in order to establish predictions relative to the characteristics of people who consult or do not consult film critics.

Knowledge

While a negative relationship between knowledge and information search has been observed in many studies (Anderson, Engledow and Becker 1979; Moore and Lehmann 1980; Newman and Staelin 1972), in other studies the observed relationship was positive (Brucks 1985; Johnson and Russo 1984). These contradictory results have been explained by the fact that although knowledge may facilitate information search and processing, at the same time it renders information less necessary (Brucks 1985). In some studies (Bettman and Park 1980; Johnson and Russo 1984), a curvilinear (U-inverted) relationship between knowledge and information search has been reported, which appears to reconcile the preceding contradictions. At low and intermediary levels, knowledge would have a positive impact on information search because it facilitates processing, but at superior knowledge levels the acquisition of additional information would be less useful. The first research hypothesis is based on this logic.

H₁: There is a curvilinear (U-inverted) relationship between movie goers’ knowledge about cinema and the extent of their consultation of film critics.

Susceptibility to Social Influence

A consumer’s susceptibility to social influence can be defined as his or her receptivity to others’ opinions with regard to products and services to buy and consume. Consumers highly susceptible to social influence will seek to obtain from important others information about products and services in order to maintain or improve their social image (Bearden, Netemeyer and Teel 1989). It is useful to distinguish normative and informative social influence (Deutsch and Gerard 1955). Susceptibility to normative social influence is concerned with a person’s desire to conform to others’ expectations. Informational social influence on the other hand is founded on a person’s desire to obtain information from others. It is proposed that consumer susceptibility to the two types of social influence is positively linked to the consultation of film critics. The second hypothesis follows from this general proposition.

H₂: There is a positive relationship between movie goers’ susceptibility to scial influenceBboth informational and normativeBwith regard to cinema and the extent of their consultation of film critics.

Self-esteem

A negative relationship has been observed between self-esteem (i.e., a person’s generalized self-confidence) and the propensity to be persuaded easily (Janis 1954). Duncan and Oshalvsky (1982) have found that information search is significantly greater among consumers having a bad opinion of their abilities to judge product quality. Therefore, it is proposed that movie goers who consult film critics have a more negative image of themselves than those who do not.

H₃: There is a negative relationship between movie goers’ self-esteem and the extent of their consultation of film critics.

Involvement

There have been few studies examining the relationship between product involvement and information search. Bloch, Sherrell and Ridgway (1986) have obtained positive and significant correlations between personal involvement towards products (personal computers, clothes) and the propensity to search for information about these products. The relationship between involvement and information search may however depend on the nature of the product. Mittal (1989) suggests that in the case of expressive products, i.e., products where functional or physical performance is less important than symbolic value, there is no necessary relationship between involvement and information search. According to Mittal (1989), for expressive products it is the symbolic meanings that are important rather than physical attributes. Whereas physical attributes are well suited for external information search, symbolic meanings are largely idiosyncratic. Cinema being an expressive product in the sense conveyed by Mittal (1989), one can predict that personal involvement with regard to cinema has no impact on information search. The fourth and last research hypothesis is based on this idea.

H₄: There is no relationship between movie goers’ personal involvement with cinema and the extent of their consultation of film critics.

METHOD

To test the research hypotheses, a survey was conducted with a sample of 120 amateurs of cinema. The sample was composed of university students. The use of students in this research seems acceptable since this population belongs to the age group (18-24 years) whose probability to go to the movies at least once a month is higher than the average (Dortch 1996).

The questionnaire consisted in scales aimed at measuring the concepts of interest, namely the extent of consultation of film critics, subjective knowledge about cinema, susceptibility to social influence, self-esteem and personal involvement with cinema. In addition, the questionnaire included measures of cinema frequentation, money spending for cinema, appreciation of 13 different movie genres (action, adventure, biography, comedy, drama, history, horror, police, science fiction, spectacle, spy, thriller, war) as well as socio-demographic questions (age, income, sex, domain of study and number of completed years in school).

Measures

The extent of consultation of film critics was measured using four items for which subjects had to indicate their level of agreement on a seven-point bipolar scale. These are original items since it was not possible to find a scale for this purpose. A typical item is: "Most of the time, I read crtiques of movies in newspapers or magazines".

Knowledge about cinema was measured using five original items. A typical item is: "I have more knowledge about cinema than others". Again, subjects indicated their level of agreement on seven-point bipolar scales. Although this is a subjective measure of knowledge, it must be noted that objective knowledge and subjective knowledge are generally positively correlated (see e.g., Brucks 1985). Some researchers (Park and Lessig 1981) argue that subjective measures of knowledge may in fact be preferable because they are indicators of one’s objective knowledge as well as one’s self-confidence. Thus, a lack of self-confidence may lead to more information search, independently of the true level of knowledge.

Susceptibility to social influence was measured by means of eleven items for which subjects indicated their agreement on seven-point bipolar scales. These items were taken from a scale developed by Bearden, Netemeyer and Teel (1989) and adapted to the context of cinema. Four of the items aim at measuring susceptibility to informational social influence (e.g., "To make sure I choose the right movies, I often observe what movies others see") and the other seven items are concerned with normative social influence (e.g., "I rarely go to see new movies unless I am sure my friends approve of them").

Seven items are used to assess self-esteem. They are taken from the scale developed by Rosenberg (1965). A typical item is: "In general, I tend to think of myself as a failure". Subjects indicated their level of agreement with all items by means of seven-point bipolar scales.

Finally, personal involvement in cinema was measured by means of ten seven-point bipolar scales borrowed from Zaichkowsky’s (1985) involvement scale. Examples of scales are: To me cinema is: unimportant/important, irrelevant/relevant, superflous/vital, etc.

TABLE 1

SCALE RELIABILITIES

RESULTS

Sample Description

Survey participants are 120 graduate students in business administration in a French-Canadian university. The sample comprises about the same number of men (50.8%) and women (49.2%) with a mean age of 25 years. On average, they go to the movies 30 times a year and spend 164 dollars (Canadian) a year for movie tickets. This higher than average level of spending for this category of product (Dortch 1996) may be explained by the fact that students are great consumers of cinema and also by the presence of a movie theatre on campus. The variance in movie attendance is high: 15% of respondents are small users (once a month or less) and 25% are heavy users (once a week or more). Knowledge about cinema also varies greatly: on the subjective knowledge scale (minimum=5, maximum=35), 30% of respondents have a score of 12 or less, 40% a score comprised between 13 and 23 and 30% a score of 24 or more (mean=17.63; standard deviation=6.98).

Quality of the Measures

Scale reliability was assessed with Cronbach’s alpha coefficient (Nunnally 1978). The results are displayed in Table 1. As can be seen, reliability coefficients go from satisfactory (susceptibility to informational social influence, a=0.62) to excellent (knowledge about cinema, a=0.91). Factor analyses of scale items resulted in a dominant factor in each case. Accordingly, all measures correspond to the sum of items composing the scales.

Some results allow the assessment of the validity of the knowledge and personal involvement scales. Thus, knowledge about cinema is positively correlated with the total number of movie genres that are appreciated (r=0.22, p<0.05). Knowledge about and personal involvement with cinema are positively correlated with annual movie attendance (r=0.55, p<0.001; r=0.31, p<0.001, respectively) and annua spending for movie tickets (r=0.52, p<0.001; r=0.28, p=0.002, respectively). These correlations are consistent with what one would expect and, as such, provide empirical evidence of construct validity.

Model Specification

In order to test the research hypotheses, a multiple regression model was estimated. This model comprises the consultation of film critics (Critics) as dependent variable and knowledge about cinema (Know), susceptibility to normative (Norm) and informational (Inform) social influence, self-esteem (Esteem) and personal involvement with cinema (Involv) as independent variables. Since a curvilinear relationship is predicted between knowledge and consultation of film critics (H₁), the regression model includes both linear and quadratic coefficients associated with the knowledge variable. As recommended by Neter, Wasserman and Kutner (1985), the independent variables were expressed as deviations from their mean in order to attenuate the inherent multicollinearity associated with polynomial regression models.

The regression model is as follows (all independent variables are centered around their mean):

Critics_i = b₀ + b₁ Know_i + b₂ (Know_i)² + b₃ Norm_i + b₄ Inform_i + b₅ Esteem_i + b₆ Involv_i + e_i

Multicollinearity Checks

Correlation coefficients between the independent variables are displayed in Table 2. In general, correlations are low. The highest correlation is that between knowledge and involvement (r=0.58), and it is inferior to the model multiple correlation coefficient (R=0.75) (Maddala 1977).

Model Estimation and Test

The regression model estimation results are presented in Table 3. The model is statistically significant (F=27.58; p<0.001) and altogether the independent variables explain 57% of the total variance (adjusted R²).

Consistent with H₁, a statistically significant curvilinear relationship is observed between knowledge about cinema and consultation of film critics. The linear regression coefficient is positive (b₁=0.47) and significant (t=7.45; 113 df; p<0.001). The quadratic coefficient is negative (b₂=- 0.02) and also statistically significant (t=- 3.13; 113 df; p<0.001). This U-inverted relationship supports H_1. The graph of the relationship is displayed in Figure 1. It can be seen that the consultation of film critics generally increases as knowledge increases. However, at some point the extent of consultation declines sharply.

TABLE 2

CORRELATION MATRIX OF THE INDEPENDENT VARIABLES

TABLE 3

REGRESSION OF CONSULTATION OF FILM CRITICS (CRITICS) ON KNOWLEDGE (KNOW), SUSCEPTIBILITY TO NORMATIVE (NORM) AND INFORMATIONAL (INFORM) SOCIAL INFLUENCE, SELF-ESTEEM (ESTEEM) AND PERSONAL INVOLVEMENT (INVOLV)

The regression coefficient associated with the susceptibility to normative social influence variable is positive (b₃=0.28) and statistically significant (t=4.42; 113 df; p<0.001). This is also the case for that associated with informational social influence (b₄=0.51; t=7.11; 113 df; p<0.001). H₂ is therefore supported: as movie goers’ susceptibility to social influence (normative and informational) increases, so does the extent of their consultation of film critics.

The regression coefficient associated with self-esteem is negative (b₅=- 0.18) and statistically significant (t=- 4.76; p<0.001). These results support H₃: the more positive moie goers’ self-esteem the lesser the extent of their consultation of film critics.

The regression coefficient associated with personal involvement with cinema is negative (b₆=- 0.20) and statistically significant (t=- 3.72; p<0.001), which is not consistent with H₄. In contradiction with what was predicted (i.e., no impact of involvement on consultation), it seems that the more movie goers are personally involved with cinema, the lesser the extent of their consultation of film critics.

Relative Importance of Explanatory Variables

Since the independent variables are not strongly correlated, one can examine the standardized regression coefficients to assess the relative impact of the independent variables in the context of the estimated model (Hair, Anderson, Tatham and Black 1995). As can be seen in Table 3, the standardized coefficient associated with knowledge (0.61) is larger than those associated with the other variables, which implies that this variable’s contribution to explained variance is greater. In comparison with self-esteem which is a generalized personality trait, knowledge as measured in this study is directly linked to cinema matters. This greater specificity probably explains why it contributes more importantly to explaining such specific a behavior as the consultation of film critics (Ajzen and Fishbein 1977).

FIGURE 1

GRAPH OF THE RELATIONSHIP BETWEEN KNOWLEDGE ABOUT CINEMA AND CONSULTATION OF FILM CRITICS

It is interesting to note that movie goers’ susceptibility to informational social influence (0.49) has a greater impact on consultation than their susceptibility to normative social influence (0.30). This implies that the motives of movie amateurs with regard to the consultation of film critics are more centered around information acquisition than a need to conform to others’ expectations.

The rather small importance of involvement in the model appears to be symptomatic of the difficulty to identify a general relationship between personal involvement and various consumer behaviors. Personal involvement is generally associated with a more intensive search for information (Bloch, Sherrell and Ridgway 1986; Zaichkowsky 1985) partly because involvement seems to be naturally linked to consumption situations perceived as more risky, more difficult (Laurent and Kapferer 1985), situations where it is necessary to obtain more information in order to reduce uncertainty. However, it is possible to be highly involved towards a product category without perceiving a buying situation as risky or difficult. Cinema is a relevant example.

DISCUSSION

Three out of four research hypotheses put forward in this article have received empirical support. As predicted, a curvilinear relationship between subjective knowledge about cinema and the intensity of consultation of film critics has been observed. This relationship is consistent with what other researchers have found in different consumption domains (Bettman and Park 1980; Johnson and Russo 1984). While the impact of knowledge on the consultation of film critics is positive and significant, there is a significant decrease in the extent of consultation at higher levels of knowledge. It must be noted that we would have concluded to a general positive effect of knowledge on the consultation of film critics if only a linear coefficient had been included in the regression model. This shows the importance of considering more complex explanatory models when interest centers on examining the relationship between knowledge and information search.

The results obtained in this study with regard to the impact of susceptibility to social influence are quite clear: the more amateurs of cinema are preoccupied with the opinions and expectations of others in cinema matters, the greater the extent of their consultation of film critics. The two dimensions that define the concept of susceptibility to social influence, namely the normative and informational imensions, have both a statistically significant impact on the consultation of film critics. However, an examination of the standardized regression coefficients indicates that the informational dimension has more weight. This result cannot be attributed to differences in scale reliability since the normative social influence scale is actually more reliable (see Table 1). It seems logical to conclude that the consultation of film critics is more associated with one’s desire to use others as sources of information than with one’s need to conform to some norms defined by important others with regard to cinema.

The relationship between self-esteem and consultation of film critics is negative, as predicted by H₃. The more negative movie goers’ perceptions of themselves, the more they use film critics to gather information about movies. This is an interesting result since it concerns a relationship between a personality variable and a very specific behavior. In a study of the impact of self-confidence and anxiety on information search, Locander and Hermann (1979) have observed a statistically significant relationship between a specific measure of self-confidence (e.g., "I would/would not be confident of my ability to pick the best buy from available brands") and the tendency to look for information from six different sources. However, in that study a generalized measure of self-esteem from a scale developed by Coopersmith (1967) had not significant effect on information search. Locander and Hermann (1979) explain this result by arguing that a personality trait should not be related to specific behaviors. Why then is it the case that in the present research such a relationship was uncovered ? It is not possible to make any judgement on the quality of the self-esteem measure used by Locander and Hermann (1979) since nothing is mentioned in the article about its reliability. A possible explanation may be found in the different nature of products under research. The products used as stimuli in the Locander and Hermann (1979) study were utilitarian products (paper towels, after-shave/cologne, toaster, lawn mower, stereo). These types of product are characterized by functional attributes about which it is possible to obtain relatively objective information. In that case, it is hard to see why one’s self-image should have any impact on information search. However, in the case of a product like cinema it is the symbolic meanings which are of primary importance and, as Mittal (1989) points out, these meanings are idiosyncratic. One’s self-image would tend to be associated with one’s perception of his or her capacity to form a personal judgement about a movie and the consequent need to obtain information from expert sources. These explanations are necessarily speculative and need to be corroborated in future research.

The most surprising result of this research is the observation of a negative relationship between personal involvement with cinema and the consultation of film critics. H₄ is founded on Mittal’s (1989) argument that the relationship between involvement and information search is moderated by the nature of the product. More precisely, Mittal (1989) argues that the theoretical positive relationship between involvement and search should apply only to utilitarian (or functional) products for which attribute information can be objectively searched. For expressive products like cinema, the symbolic interpretations matter more than attributes and these interpretations, because they are basically personal, are less susceptible to be formed through information search. To test this idea, Mittal (1989) compared indices of information search associated with expressive products with indices associated with functional products in situations of high product involvement and found that search was indeed less intensive for expressive products than for functional products. However Mittal (1989) did not make comparisons between low and highly involved consumers (a limitation he admits) and could not consequently examine the relationship between involvement and information search. What we observe in the present research where personal involvement is measured is not only that there i no positive impact of involvement on the consultation of film critics, but that this impact is negative and statistically significant. It would be necessary to corroborate this finding in future studies using different expressive products as stimuli. But in the meantime, it is very interesting to observe that the nature of products may change the direction of the relationship between involvement and information search.

LIMITATIONS AND DIRECTIONS FOR FUTURE RESEARCH

This study has methodological limitations that should be eliminated in future research. Firstly, it is a correlational study. Thus the effects of the individual variables (knowledge, self-esteem, etc.) that were observed apply to the consultation of film critics reported by the survey participants and this may be different from true consultation. It is for instance possible that some participants overestimated the intensity of their consultation to protect their self-image. That might explain some results, such as the negative relationship between self-esteem and consultation of film critics. It would be important in future studies to consider using observational measures of consultation of film critics. One could for example use the method of information display boards (see Jacoby et al. 1976) to obtain a behavioral measure of information acquisition from film critics in situations of movie evaluation or choice.

A second methodological limitation concerns the knowledge measure used in this study which was defined around subjective rather than objective knowledge. It would be relevant in future studies to try to verify if the U-inverted relationship between knowledge and consultation of film critics still holds when knowledge is measured objectively. The work of Brucks (1985) might be useful for this purpose.

A third limitation concerns the sample of amateurs of cinema having participated in the survey. While students represent a relevant segment for a product like cinema, movie goers can be found in other groups of society too. One negative consequence associated with the use of students is that other individual variables which are important to predict the extent of information search such as age, education and income (see Newman 1977) do not have enough variance to allow an estimation of their impact. It would be important that future studies aimed at explaining the extent of consultation of film critics use more representative samples of the population of amateurs of cinema or samples that show sufficient heterogeneity to allow the estimation of the impact of other relevant variables.

An important conceptual limitation of this research concerns the fact that it is restricted to an examination of a few individual factors to explain the extent of consultation of film critics by movie goers. Other variables such advertising, the reputation of the actors and the film director, consumer preferences, the difficulty in obtaining information about movies, etc. can have a significant impact on the intensity of consultation. The present research has focused on explaining a general tendency to consult film critics. Future research should conceive of the problem from more specific points of view, for example by studying the factors that explain the consultation of films critics for certain types of movie in certain situations.

REFERENCES

Ajzen, I. and Fishbein, M. (1977). Attitude-behavior relations: A theoretical analysis and review of empirical research. Psychological Bulletin, 84, 888-18.

Anderson, R.D., Engledow, J.L. and Becker, H. (1979). Evaluating the relationship among attitude toward business, product satisfaction, experience, and search effort. Journal of Marketing Research, 16, 394-400.

Bearden, W.O., Netemeyer, R.G. and Teel, J.E. (1989). Measurement of consumer susceptibility to interpersonal influence. Journal of Consumer Research, 15, 473-481.

Bettman, J.R. and Park, C.W. (1980). Effects of prior knowledge and experience and phase of the choice process on consumer decision processes. Journal of Consumer Research, 7, 234-248.

Bloch, P.H., Sherrell, D.L. and Ridgway, N.M. (1986). Consumer search: An extended framework. Journal of Consumer Research, 13, 119-126.

Boor, M. (1992). Relationships among ratings of motion pictures by viewers and six professional movie critics. Psychological Reports, 70, 1011-1021.

Brucks, M. (1985). The effects of product class knowledge on information search behavior. Journal of Consumer Research, 12, 1-15.

Coopersmith, S. (1967). The antecedents of self-esteem. San Francisco: W.H. Freeman & Co.

Deutsch, M. and Gerard, H.B. (1955). A study of normative and informational influence upon individual judgment. Journal of Abnormal and Social Psychology, 7, 1-15.

Dortch, S. (1996). Going to the movies. American Demographics, 18, 4-7, 22-23.

Duncan, C.P. and Olshavsky, R.W. (1982). External search: The role of consumer beliefs. Journal of Marketing Research, 29, 32-43.

Eliashberg, J. and Shugan, S. (1997). Film critics: Influencers or predictors ? Journal of Marketing, 61, 68-78.

Faber, R. J. and O’Guinn, T.C. (1984). Effect of media advertising and other sources on movie selection. Journalism Quarterly, 61, 317-377.

Farber, S. (1976). The power of critics. The American Scholar, 45, 419-423.

Hair, J.F., Anderson, R.E., Tatham, R.L. and Black, W.C. (1995). Multivariate data analysis, Fourth Edition. Upper Saddle River, NJ: Prentice-Hall.

Hugstad, P., Taylor, J.W. and Bruce, G.D. (1987). The effects of social class and perceived risk on consumer information search. Journal of Consumer Marketing, 4, 41-46.

Jacoby, J., Chesnut, R.W., Weigl, K.C. and Fisher, W. (1976). Pre-purchase information acquisition: description of a process methodology, research paradigm, and pilot investigation. In B.B. Anderson (Ed.), Advances in consumer research, Vol. III (pp. 306-314). Chicago: Association for Consumer Research.

Janis, I.L. (1954). Personality correlates of susceptibility to persuasion. Journal of Personality, 22, 504-518.

Johnson, E.J. and Russo, J.E. (1984). Product familiarity and learning new information. Journal of Consumer Research, 11, 542-550.

Laurent, G. and Kapferer, J.-N. (1985). Measuring consumer involvement profiles. Journal of Marketing Research, 22, 41-53.

Lavidge, R.J. and Steiner, G.A. (1961). A model for predictive measurements of advertising effectiveness. Journal of Marketing, 25, 59-62.

Locander, W.B. and Hermann, P.W. (1979). The effect of self-confidence and anxiety on information seeking in consumer risk reduction. Journal of Marketing Research, 16, 268-274.

Maddala, G.S. (1977). Econometrics, New York: McGraw-Hill.

Mittal,, B. (1989). Must consumer involvement always imply more information search ? In T.K. Srull (Ed.), Advances in consumer research, Vol. XVI (pp. 167-172). Provo, UT: Association for Consumer Research.

Moore, W.L. and Lehmann, D.R. (1980). Individual differences in search behavior for a nondurable. Journal of Consumer Research, 7, 296-307.

Neter, J., Wasserman, W. and Kutner, M.H. (1985). Applied linear statistical models, Second Edition. Homewood, IL: Richard D. Irwin.

Newman, J.W. (1977). Consumer external search: Amount and determinants. In A.G. Woodside, J.N. Sheth and P.D. Bennett (Eds.), Consumer and industrial buying behavior (pp. 79-94). New York: North-Holland.

Newman, J.W. and Staelin, R. (1972). Prepurchase information seeking for new cars and major household appliances. Journal of Marketing Research, 9, 249-257.

Nunnally, J.C. (1978). Psychometric theory. Second Edition, New York: McGraw-Hill.

Park, C.W. and Lessig, V.P. (1981). Familiarity and its impacts on consumer decision biases and heuristics. Journal of Consumer Research, 8, 223-230.

Reddy, S., Swaminathan, V. and Motley, C. (1996). Hits on Broadway: The role of critic’s reviews in the success of an experimental good. In E. A. Blair and W. A. Kamakura (Eds.), Proceedings of the 1996 AMA winter marketing educators’ conference (pp. 313-324). Chicago, IL: American Marketing Association.

Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press.

Udel, J.G. (1964). A new approach to consumer motivation. Journal of Retailing, 40, 6-10.

VTzina, R. (1997). One of the best movies of the year !: An experiment on the influence of critics’ recommendations on consumers. In D. Arnott, D. Doyle, V. Shaw and R. Wensley (Eds.), Proceedings of the XXVIth annual conference of the European Marketing Academy, Vol. 3 (pp. 1300-1312). Warwick, Great Britain.

Wyatt, R. O. and Badger, D. P. (1984). How reviews affect interest in and evaluation of films. Journalism Quarterly, 61, 874-878.

Zaichkowsky, J.L. (1985). Measuring the involvement construct. Journal of Consumer Research, 12, 341-352.

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