Youth Viewers of Pakistan Television (PTV) and the Enculturation Model of the Islamization Process: Towards Exploring Some Empirical Basics Fazal R. Khan Introduction Although it was introduced only a little Over two decades ago, Pakistan Television (PTV) seems to have made tremendous progress in terms of the standards set by developing countries. PTV now covers 86.4 percent of the total population? In a national sample, over 50 percent of all men aged twenty- one and above were found to watch TV daily or often, while the rest watched less frequently or never.2 Females, despite the lack of concrete data, are likely to be even higher consumers of TV fare due to their homebound social ~ t a tus .~ Similarly, it appears that both males and females under twenty-one years of age watch even more television. This precipitate pervasiveness of TV in a developing society like Pakistan gives the entertainment TV or entertainment video element of the indigenous culture industry vast potentialities in any serious strategy designed to further social development and to socialize audience members into family, community, and sociopolitical norms within the system at the microindividual, interpersonal subsystem, and macrosocietal levels. Unfortunately, there have been very few theoretically based empirical studies done on any aspect of PTV transmissions Fazal R. Khan is currently the chairman of the Department of Journalism and Mass Communication at Gomal University, NWFP, Pakistan. 'H. Rashid, ed., PTV Basics 1989 (Karachi: Pakistan Television Co., Ltd., 1989). 25. ZOrient Advertising (Pvt.) Ltd., Pukisiun Adverrising Scene Po (Karachi: Market Research Department, Orient Advertising, 1990), 66. 31n fact, a secondary stepwise regression analysis of the present data revealed that females were more likely to watch TV programs more frequently and for longer periods of time. LU at the communicator and/or audience end of the communications tran~actions.~ Nevertheless, Pakistan's recent legislative adoption of the Shari'ah (Islamic law) into its legal system seems to make the need for conducting such research even more urgent, for EyTV's role in the Islamization of Pakistan now has to be defined. In the past, successive Pakistani administrations have sought to Islamize the country by imposing legislative enactments in a synchronic, top-down, and information-centric mode. No concern for the development of media- based strategies designed for the Islamic enculturation of Pakistani society at the extralegal or human levels was ever voiced. If such a strategy is to be adopted, however, a major role will have to be given to the socialization, enculturation, and the representational realism potentials of the indigenous video-media industry. Thus, PTV will emerge as its predominant component. Due to the array of pertinent mass communication theoretical traditions developed in the West, which have recently been posited as having important bearings on the Islamic enculturation of a given ~ociety,~ it is possible to develop outlines of what may be called an entertainment video or enculturation model of the process of Islamization (see figure 1). An Enculturation Model of the Islamization Process A closer look at figure 1 shows that the posited model is not a simple short-term, cross-sectional conditioning model. It is rather a stimulusaganism- response, long-term model of intended effects predicated on creating a purposive mass media environment of communications transactions in a given society. In other words, it has much in common with McQuail's process model of collective consciousness forming and reality structuration. More specifically, the present model utilizes wisdom from the society- directed Qur'anic concept of a1 baliigh a1 mubin (unambiguous communication). It is assumed that unambiguous communication cannot occur in a society in which considerable "noise" is generated via other channels, 'With the exception of F. Khan, "Violence on Television in Asia: A Case of Pakistan," (Asian Mass Communication Research and Information Centre, Singapore, 1989, mimeographed), which attempts to profile some aspects of PTV's message system, most other studies have not gone beyond describing the frequencies of FTV use. Prominent among these seem to be the national data reported in Orient Advertising (Pvt.) W., op. cit. 5F. Khan, "Entertainment Video and the Process of Islamization in Pakistan: Theoretical Perspectives on a Eolicy Imperative," American Journal of Islamic Social Sciences 8, no. 2 (September 1991): 289-306. 6D. McQuail, Mass Communication Zheory: An Znrroduction (Beverly Hills, CA: Sage, 1984), 180-211. man: xoum viewers or m s m ieievision as these aid in the cultivation of alternate conceptions of social reality. It thus goes beyond such long-outdated unidimensional dicta like “the medium is the message,” “content equals effects,” or even “a powerless media via-‘a-vis an obstinately impervious audience” by suggesting a communications transaction in a purposive mass media environment within a given society. It is this holistic mass media environment, constituting a synchronous message system of entertainment video’s symbolic world rather than just a single message, which may gradually cultivate the structures of Islamic norms, sociopolitical relationships, roles, and other elements of Islamic culture among audience members and, through them, into the general environment and vice versa. The model takes into consideration the general environment within which audience members exist as well as the elements related to their demographic and psychographic makeup and their subjective social reality. In other words, audience members are not assumed to be atomized homo mechanicus but rather homo supiens wZens7 - individuals possessing a mind and the ability to think locked in a web of intra- and extrafamily relationships and interaction patterns in addition to having a specific orientation(s) towards and perceptions of society. Specifically, the model contains two sets of components: column “a” and column “b.” The first column deals with stimuli development and infusion strategies within the wider societal and cultural environment, while the latter column stresses programmatic studies of audience members interfacing with primary stimuli within the context of the wider sociocultural and societal environment. This latter column also includes the nature and types of overall effects on them and the implications for an ongoing process of primary stimulus development and refinement. The concept of a primary stimulus implies the Islamic Socialization Potential (ISP) of the message system encoded in an entertainment format. The major elements under column “a” are as follows: a) a talent pool of core Islamic cultural elites who not only mind import structures regulating global cultural imports* but whose cultural productions a€so help influence the general environment and all aspects of the culture industry, including the National Institute of Mass Communication Policy and Research (NIMCPR); ’H. Mendelsohn, “Mind, Affect, and Action: Construction Theory and Media Effects Dialectic,” in Mass Communication and Political Information Processing, ed. S. Kraus (New Jersey: LEA, lW), 43. Wlobal cultural imports include, among others, satellite- and nonsatellite-based entertainment, commercial, and informational fares. Attention must be paid to commercial fares, for these are likely to cut at the very root of economic self-discipline by creating contra- Islamic, if nat un-Islamic, consumerism and Islamic acculturation. 22 The Amencan Journal or isiamic ~ W I U O L I G ~ ~ ~ ~ ~ 7 . L b) the phrase ”general environment’’ denotes a composite of elements which constitute objective social reality within a given ~ociety.~ Chief among these a h the structural and institutional, and the legislative and sociocultural elements. Also included are the culture industry’s elements of print and electronic media as well as the various agencies of education and socialization other than the entertainment video industry; c) the PTV-led entertainment video industry also contains cinema and VCR entertainment fare. The important factor here is the availability of video media and the array of choice as regards entertainment content rich in ISP constituting video-media’s symbolic reality for audience members; and d) NIMCPR‘s research and development apparatus will be made responsible for two basic functions: 1) developing an ISPll index in conjunction with the entertahment video industry through a continuing program of message system analysis and 2) forging a closer fit between the message system and the cultivation and enculturation effects by means of an ongoing program of cultivation and enculturation effects analysis. In column “b,” the main elements are: a) audience members; b) communication processes and perspectives which determine the audiences’ nexus with the primary and secondary stimuli emanating respectively from the entertainment video industry and the general environment; and c) the ultimate effects which build up accretively and over time. Theoretical perspectives which may come into play at linkages 6, 7, and 8 include mainstreaming and cultivation effects of social reality research,’2 agenda-setting effects of information distribution re~earch:~ opinion formulation and the spiral of silence’s as well as the pluralistic ignorance research’s14 distribution effects, ideology formation, the marginalization effects of the signification and culturalist researchl5 media images, gratifications 9H. Adoni and S. Mane, “Media and the Social Construction of Reality: Toward an Integration of Theory and Research,” Communication Research 11, no. 3 (1984): 325. losee F. Khan, op. cit., 17-9 for the institutional role and structure of NIMCPR. ”See ibid., 19-28 for a detailed conceptual explication of the ISP index. I2G. Gerbner et al., “he Mainstreaming of America: Violence Profile no. 11,” Journal of Communiaation 29, no. 3 (1979): 177-96. I3M. McCombs, ”The Agenda-setting Approach,” in The Handbook of Political Communication, eds. D. Nimmo and K. Sanders (Beverly Hills, CA Sage, 1981): 12141. Further refinements of the concept appear in S. Iyengar and D. Kinder, News Zhd Matters (Chicago: University of Chicago Press, 1987). “E. Noelle-Neuman, ’The Spiral of Silence: A Theory of Public Opinion,” Journal of Communication 24 (W4): 43-51. 15T. Bennett, “Media, ‘Reality,’ Signification,” in Culture, Society and the Media, ed. M. Gurevitch et al. (London: Methuen, 1982), 285-308. See also J. Fiske and J. Hartley, Reuding Television (London: Methuen, lq8). effects of media use and effects tradition of communication research,16 and prosocial and antisocial role socialization effects of media socialization re~earch?~ Given the overall wisdom of the above theoretical perspectives at llnkages 6, 7, and 8 as discussed at some length elsewhereta a general theoretical expectation according to the posited model wuld be that a sustained infusion of symbolically stable, systematic, and culturally syncretizing messages rich in ISP content into a society featuring a supportive media and general environment is likely to lead, over time, to the Islamic enculturation of audience members and their general environment. Clearly, this theoretical statement is predicated on a rich research agenda requiring a sustained program of institutionalized research at linkages 6, 7, and 8 within the broad framework of the posited model. Subsumed among the model’s research prescriptions is the basic need for understanding how the audience members interface with the present culture industry and their general environment. Specifically, any enculturation strategy based upon the posited model is tied to generating empirical knowledge about the sociocommunication, demographic, and psychographic makeup of existing PTV entertainment viewers-an ultimate nucleus of any strategy.” In accordance with these understandings, then, the remainder of this paper seeks to piece together those aspects of the sociocommunication, psychographic, and demographic profiles of youth which are relevant to hkages 6 and 7 of the enculturation model. The category of youth has been chosen because this is the group most likely to be the heaviest consumer of FTV 16E. Katz, J. Blumler, and M. Gurevitch, ”Utilization of Mass Communication by the Individual,” in f i e Uses of Mass Communimtion, eds. J. Blumler and E. Katz, 9th ed. (toradon: Sage, 1974). See also J. McLeod, C. Bybee, and J. D u d , “Evaluating Media Performance by Gratification Sought and Received,” Jownalism Quarterly, 59 (1982): 3-13; and G. Kosiciki and J. McLeod, “Media Information Strategies and Media Images: Effects on Learning, Interest and participation” (paper delivered at the Midwest Political Science Association, Chicago, Illinois, April 1989). 17J. Rushton, “Television and Pro-Social Behavior: in Television and Social Behavior, vol. 2: Technical Reviews, NIMH; see also W. Brown, “Prosocial Effects of Entertainment Television in India,” Asian J o u m l of Communication 1 (1990): 113-35. 18F. Khan, op. cit. 191n fact, symmetrical arrows point to the salience given by the enculturation model to the audience members, thereby implying a diachronic strategy of symbolic d i t y infusion into the system. A diachronic strategy implies an understanding of the target audience by way of a cyclic process in which information seeking precedes stimulus construction and its subsequent infusion into the system. See, for example, S. Rahim, “Determinants of Diachronic Information Seeking among Communication Intermediaries,” Asian Journal of Communication 1, no. 1 (1990): 98-112, where diachronic communication in its limited application has been discussed. L-r programs (viewers between the ages of twenty-one and twenty-four view PTV programs with the highest degnx of regularity and, in comparison to other age groups, the smallest proportion of nonviewers Mls into this categoryzo) and because evidence has been overwhelming and mounting, at least in the West, that youth are the most vulnerable to mass media and cultivation effects. For example, it has been found that younger viewers born into a TV world are more imbued with its depiction of people and therefore are less likely to express an independent view of reality. Also, cultivation effects are stronger for adolescents.z1 Fnrm the standpoint of the Islamization enculturation model, therefore, knowing about the communication behaviors of this group is more central, for it is likely to be more susceptible to the symbolic world of PTV/entertainment video, thereby constituting an early nucleus of any enculturation effort. This study was specifically designed to address the issues of whether there are meaningful patterns of PTV viewing among youth and, if there are, how they might relate to other sociocommunication, demographic, and psychographic variables. In other words, how might we define PTV viewers’ interfacing with aspects of a given society’s general environment and entertainment video? Method Sample Data were collected through personal interviews from a stratified systematic sample of three hundred six students enrolled at the affiliated and constituent colleges and departments of Gomal University, Dera Ismail Khan, Pakistan, during February and March 1987.z2 Completed interviews constituted 94.2 percent of the original sample. Male and female graduate students in an advanced level research method class at the university conducted the interviews under the close supervision of the staff of the university’s Mass Communication Research Center. %ee Orient Advertising (FW.) Ltd., op. cit. ”G. Gerbner and N. Signorielli, “Women and Minorities in Television Drama, 1%9-1978,” (Annenberg School of Communication, Philadelphia, Pennsylvania, l w , mimeographed), 29. See also G, Gerbner et al., “Aging with Television: Images on Television Drama and Conceptions of Social Reality,” Journal of Comnunication 30, 110. 1 (1980): 37-47; and G. Gerbner et al., “The Mainstreaming of America: Violence Profile No. 11,” op. cit. ZfAttendance rosters formed the sample frame and stratification was done on sex and class levels. The sampling interval equalled ‘5” in a random-start systematic sampling procedure. The completed interviews resulted in the following sample population: 47.4 percent males and 52.6 percent females ranging in age from fifteen to thirty-two; 46.7 percent were twenty years old or less (mode = 18 years) and 53.1 percent were over twenty years old (mode = 23 years); the mean age was 20.9 years; and, as regards education, 46.0 percent were graduate students while the rest were undergmduates. The sample was somewhat skewed towards the lower income end, with 59.8 percent of the valid cases reporting an annual household income of twenty thousand rupees or less. The rest had an annual household income of over twenty thousand rupees.23 Criterion Groups: Viewer Types The fmquency of this population's exposure to fourteen non-news television programs, measured on a four-point Likert-type scale, was factor analyzed using a principal component solution with an oblique A minimum eigenvalue of greater than or equal to one criterion, supported by a Scree test, yielded three distinctly interpretable factors. Taken together, these factors accounted for 57.0 percent of the total variance in the items (see table l).25 Factor 1 contained all of the religious programs, factor 2 comprised domestically produced entertainment programs, and factor 3 had a clustering of four programs (two were imports from the UNted States and the other two, although produced in Pakistan, were heavily dependent on Western content). Factor loadings on each of the factors genedly tended to be .6 or better. On the basis of factor loadings and the program types, the three factors were labelled mpectively as a) religious TV, b) domestic entertainment, and c) cosmopolitan entertainment. P r o g m viewing mean indices developed on the basis of these factors had Cronbach alpha d u e s of .82 (religious TV), .62 (domestic entertainment), and .69 (cosmopolitan entertainment). Three types of PTV viewers were then defined by cross-comParing the mpondents' scores on each of the three indices. The religious TV group comprised all viewers (33.1 percent) whose scores on the religious TV index were greater than or equal to their scores on both the cosmopolitan and the domestic entertainment indices. The cosmopolitan 23The group with an annual income of over twenty thousand rupees was bimodal, with the modes being the categories of twenty thousand to twnty-five thousand rupees and over twenty-five thousand rupees respectively. 24Rmpo&nts we= asked to indicate their fresuency of exposuxt to a list of PTV programs according to the hllciwing four-point scale: "1" (nevex watch), "2" (rarely watch), '3" (sometimes watch), and "4" (frequently watch). 2sTHo items, viz. exposure to TV quiz and TV health programs that tended to load about equivalentiy on multiple factors, were discarded in order to obtain a more interpretable and parsimonious solution. w entertainment group comprised all viewers (12.3 percent) whose scores on the cosmopolitan index were greater than or equal to their scores on both the religious TV as well as the domestic entertainment indices. The domestic entertainment group, though defined conservatively, was by far the largest group (54.6 percent). This group comprised viewers whose scores on the domkstic entertainment index were greater than their scores on both the religious TV and the cosmopolitan entertainment indices. Predictor Variables Five sets of sociocommunication, demographic, and psychographic predictors were used as some of the preliminary definers of the audience members' nexus with the general environment and the video-media industry. Sets of variables included newspaper, radio, and VCR use variables; intra- and extrafamily interaction variables; some variables related to audience members' orientation towards society; and such demographic variables as sex, income, age, and education. Newspaper use variables were combined into a seven-item mean index (standardized alpha = .6) comprising Likert- type items measuring frequency and extent of exposure to national and international stories as well as editorials and interest in those national affairs reported in newspapers. Two items measured exposure to radio news and interest in national affairs covered in radio newscasts. VCR use variables were a) frequency of exposure to English-language VCR movies on a four-point scale and b) a mean index of frequency of exposure to Indian Urdu-language movies and movie songs (alpha = .68). Interpersonal interaction consisted of a four-point scale measuring the likelihood of discussing national affairs with family members and a two-item index consisting of the likelihood of discussing national affairs with friends and coworkers (standardized alpha = .66). Audience orientation to the general environment in society was measured through such psychographic variables as salience-rating of the Shari'ah and other national issues (i.e., energy problems facing the nation and issues of national defense and security). In addition, a two-item index (alpha = .63) measured audience orientation towards economic issues in their general environment (i.e., unemployment, the high cost of living, or inflation).26 z6Nempper-use items were combined into an index of national affaim in newspapers that comprised three four-point frequenq-of-expure items ranging from "1" (never read) to M4w (frequptly read) national and international stories and editorials; three five-point items measuring th& extent of reading ranging fmm "1" (not read at all) to "5" (read full story); ' Khan: Youth Viewers of Pakistan Television n Statistical Analysis The primary statistical tool employed in this study was multiple discriminant analysis. This statistical procedure simultaneously differentiates between two or more criterion groups on the basis of several predictor^.^^ In other words, unlike multiple regression, multiple discriminant analysis can specifj more than one set of relationship& which it calls functions, between independent and dependent sets of variables. Since the primary interest of the study was to differentiate among the viewer types in order to better understand them, a discriminant analysis approach was designed to treat these groups as dependent variables with sets of sociocommunication, psychographic, and demographic variables as differentiating predictors. A forward stepwise multiple discriminant solution was used in an attempt to minimize Wilks’ lambda by selecting variables for entry (one at a time) into the analysis on the basis of their discriminating power.28 Additionally, separate discriminant analyses were used for examining pairwise comparisons between viewer groups by categories of demographic variables for further specificity. 29 Results and Discussion Preliminary univariate F-tests of predictor variables by PTV viewer-types are reported in table 2. As the table indicates, seven of the fifteen tests revealed significant mean differences at the alpha level of .04 and less. Of the demographic variables, only income showed up in a bivariate relationship with viewer type. The Shari‘ah and energy issue salience of the four psychographic variables, with respective F values of 3.39 (p c .04) and 3.65 (p c .03), appeared. to differentiate among the viewer types, as did the likelihood of extrafamily discussion of national affairs (F = 3.92, p c .03), and a four-point interest item ranging from “1” (no interest) to “4” (a lot of interest) in reading about national affairs in newspapers. Orientation of issues facing the society was measured according to a four-point scale with rating options being: “l” (not important), “2” (somewhat important), “3” (important), and “4” (very important). Two items concerning the high cost of living and problems of unemployment were combined into a single index because a confirmatory factor analysis revealed them together. 27W. Klecka, Discriminant Analysis (Newbury, CA: Sage, 1980), 7. 28Wilks’ lambda was employed as a criterion for selection because of its ability to maximize heterogeneity across groups and homogeneity within groups. Tairwise comparisons for the cosmopolitan and domestic entertainment groups by categories of demographic variables were dropped because some of the smallest cells were equal to the number of predictors in the solution, thereby violating the assumption for the number of permissible predictors in a discriminant solution. 28 The American Journal of Islamic Social Sciences Y: l attention to national affairs reported in newspapers (F = 7.85, p c .OOl), exposure to radio news (F = 8.45, p < .OOl), and interest in national affairs reported on the radio (F = 5.99, p < .003). Neither the two VCR use variables nor the likelihood of intrafamily national affairs interaction attained significance. However, since bivariate relationships are always open to the possibility of spurious or suppressor effects, all of the variables were retained for the multiple discriminant solution. Table 3 provides a summary of the discriminant analysis. Theoretically, in a three-group model two discriminant functions were possible, and both attained statistical significance (F, p = .0005; F, p = .007). Eigenvalues, variances, canonical correlations, and associated significance levels are shown at the bottom of table 3. All four sets of values indicate the approximately equivalent contribution of the two functions to the overall solution. Moreover, this table has two sets of coefficients. The first set gives zero- order correlations of each variable with the tvco functions. Variables are grouped in descending order based on the size of their correlations with each function. Thus the variables repmenting the interaction between intrafamily and national affairs, Shari'ah issue salience, extrafamily and national affairs interaction, interest in radio-reported national affairs, and level of education correlate more strongly with function 1, whereas the remaining ten variables correlate relatively more strongly with function 2. The second set contains the standardized discriminant function coefficients. Similar to factor loadings in a factor matrix, each of these represents the relative contribution of its corresponding variable to the function. Of the original fifteen predictors, four (age, sex, education, and interest in national affairs reported on the radio) were eliminated by the forward selection stepwise solution as being relatively poor discriminators among the three groups. This multivariate finding is mostly in line with the univariate F-tests (see table 2). The only unexpected casualty was the item dealing with the interest in national affairs reported on the radio, which may have been pushed out by its behavioral-based kin of frequency of exposure to radio news. At any rate, these particular coefficients can be used to name the functions by idenwing their underlying dominant characteristics. It appears that function 1 is primarily composed of the Islamization-based, national affairs, and socio- communication dimensions, because both intra- and extrafamily components of sociocommunication, Islamization indicated through Shari'ah salience, orientation to other national affairs (i.e., energy issues and national defense), and exposure to radio-reported news are predominant variables. Function 2, on the other hand, is characterized by video use, economic concerns, extrafamily national affairs discussion, and exposure to radio- reported news. Of the six items loading highly on this function, four relate to economic matters and video entertainment, so it seems to be basically miin: wutn viewem or ranisran ieievision LY an economics-based national affairs and video entertainment dimension. The likelihood of an interaction between the extrafamily and national affairs variables, although loading highly on this function, does not correlate as strongly with this function as it does with function 1 (see the first set of coefficients in table 3). The two dimensions account for 53.3 percent and 47.7 percent respectively of the total variance in the three groups. Table 4 lists group centroids, which are the mean discriminant scores for each function. These are provided to show the pairs of groups discriminated by each function. The type-discrimination is determined by the two type values farthest apart on a function. As is evident in the table, function 1, representing the Islamization-based national affairs and sociocommunication dimensions, discriminates between the religious TV (+ .7l) and the domestic entertainment (- .33) types. Function 2, representing the economics-based national affairs and video entertainment dimensions, separates the cosmopolitan (+1.06) and the domestic entertainment (- .19) types. Going beyond the general dimensions discriminating among viewer types, the following detailed profiles for the three types of PTV viewers were developed using variables from the discriminant analysis. Findings from pairwise type comparisons by demographic categories have been included for further information where necessary. Religious TV Viewers This group is very likely to discuss national affairs with family members, friends, and coworkers. It considers the implementation of the Shari'ah to be the most salient national issue. It follows national affairs in the newspapers, frequently listens to radio-reported news, and is likely to dismiss national issues (i.e., defense, unemployment, inflation, and energy problems) as unimportant. In terms of sociocommunication makeup, this particular group seems to be strongly involved in interpersonal interaction more at the intrafamily, as opposed to the extrafamily, level and is thus vocally involved in public affairs. Its mass media use is generally confined to reading newspapers and listening to news reports on the radio. However, those who are aged twenty or below (i.e., adolescents) are somewhat likely to add VCR use to their mass media behavior, particularly watching VCR-recorded Urdu-language movies and songs imported from India (see table 5) . Similarly, better educated and somewhat older members of this group seem to show some secondary concern for the nation's energy and defense issues. In summary, religious TV viewers generally seem to be interpersonal communication based young enthusiasts who support IsIamization and who tend to monitor national affairs probably in Shari'ah-focused terms. They 30 The American Journal of Islamic Social Sciences 9:l uniformly dismiss the nation's economic problems as unimportant and, at best, feel lukewarm or tempered concern for issues related to energy and national defense. Domestic Entertainment Viewers Like its religious TV counterparts, this particular group is also characterized by a few universals. For example, members uniformly view the implementation of the Shari'ah as unimportant. They also pay no attention to national affairs reported in the newspapers and on the radio (see tables 3 and 5 ) . This group is further distinguished by its exposure to VCR-recorded Urdu-language movies and songs imported from India and, relative to other groups, is more likely to report higher annual household incomes. This group, with the exception of females, generally accords higher salience to issues of national defense and security. Females seem to be mainly concerned with economic problems and relatively unconcerned with issues of national defense and energy. Similarly, economic salience is confined to the low-income but better-educated category within this group. As for its sociocommunication environment, the national affairs-based intrafamily interaction is exhibited only by high-income females and by those who are over twenty years old (i.e., young adults). Relative to religious TV viewers, this group is primarily extrafamily based in its national affairs interaction. Another subgroup of some significance within the domestic entertainment category is that of adolescents, aged twenty and less, who seem to include English-language VCR movies in their entertainment diet (see table 5) . In summary, this group primarily consists of non-Shari'ah-oriented Indian movie fans who belong to relatively well-off households and pay no attention to national affairs reported in the mass media. Its male members seem freer of intrafamily influences on their national affairs orientation, which is primarily characterized by attaching salience to national defense and energy issues. Cosmopolitan Entertainment Viewers This group is similar to the religious TV group in terms of its national affairs media use behavior. But this is where the similarity ends for, unlike its counterparts, this group is given to introversion in its sociocommunication interaction. Other universals for this group include its deep uniform concern for economic issues (Le., unemployment and inflation), its use of VCRs exclusively for movies imported from Hollywood, and its predominant social location in low-income households (table 3). Further analysis (table 5) shwed that while this group generally considers Khan: Youth Viewers of Pakistan Television 31 the implementation of the Shari‘ah as unimportant, such a view is even more pronounced among its young adult members (those over twenty years of age). What this particular subgroup deems more salient are issues of national defense and economics. In summary, this p u p consists of low-income, non-Islamization-oriented, economicsconscious, public affairs introverts. Their VCR use is confined to frequent exposure to English-language VCR movies. They do, however, monitor national affairs reported in the mass media. Classification Results Table 6 gives the prediction results for the multiple discriminant solution. The two functions identified in table 3 correctly classified 53.55 percent of the cases. As the table shows, discriminant solution was most successful in classifying the religious TV (65.1 percent) and the cosmopolitan entertainment (66.7 percent) groups. Although there was some misclassification in the domestic entertainment group, the prediction rate still appeared to be better than chance. The overall correct classification rate was significantly better than chance (t = 2.4; p c .02).30 Implications These findings engage aspects of research needed at linkages 6 and 7 of the above-outlined enculturation model. However, these constitute only a minor initial step in what may be called the model’s programmatic research prescriptions. One important question which needs to be given top priority now is why are the profiles different? In other mds, what exactly is causing the difference, or how can we parsimoniously account for the difference? Is it because the three types of viewers identified above are inherently different people characterized by differences in their sociostructural location and socialization patterns? Is it because of their unequal access to social resource capital in the society, or is it something else? Specifically, is it because some kind of motivational d i f imce or human deficit may be producing different stimulus processing strategies ranging from media scanning to active processing to reflective integration? How might characteristics of their general environment explain the difference? Could it also be argued that the difference is primarily due to their different media use patterns or the characteristics of their chosen 30For the computation formula, see C. Cobb, “Patterns of Newspaper Readership among Teenagers,” Communication Research 13 (1986): 299-326. 32 The American Journal of Islamic Social Sciences 9:l media? That is, might it be due to the difkrential cultivation potential of their videmedia diet? In short, the deterministic as well as the transaction perspectives may provide explanations for the different profiles. In other words, the general environment-centric, the video-media-centric, the audience-centric, and the transaction perspectives will need to be tested. It is important that this be done, because the Western communications perspectives, their cross-cultural validation in some cases notwithstanding, at best constitute direction pointers awaiting cross-cultural validation. Of immediate interest, however, would be to investigate the differential salience accorded to the Shari'ah by the religious TV and entertainment groups. If any assertion is to be made reg- the predominance of the media stimulus influence, then the stimulus difference, viz. exposure to different media contents, will have to explain the variances in the salience accorded to the Shari'ah beyond what might be explainable through audience characteristics (i.e., demographics, psychographics, subjective reality) and characteristics of the general environment, including the influence of extra- and intrafamily interpersonal communication sources. Simultaneous controls will probably have to be included for all these in a hierarchical multiple regression design. The family influence potential of the Islamization enthusiasts and the extrafamily communication base of the Indian movie fans are instructive. Nevertheless, if stimulus influence is found, it would be imperative to profile the message system implicit in the contemporary entertainment diet, be it domestic or nondomestic in origin, available to the indigenous youth through the entertainment video industry. Of particular interest, then, would be how to examine its socialization or cultivation potentials as well as its message system. Is the entertainment video reality rich in contra-Islamization potential? How do certain aspects of the general environment support or detract from it?31 Unless an empirical base for these issues is developed, any talk of video- media reality and its enculturation effects will remain speculative at best. T h i s author has started work on developing a message system profile of FTV. It is now in the analysis stage. w 0 Cultural Imports 1 + r Figure 1: Enculturation Model of the Islamization Process (a) General Environment Social environment Socialization agencies Other media channels Legislative & other institutional structures Objective social reality 4 4 i Filters of Islamic culture industry 6 industryITV availability Content array " i Audience Members I Demographics Psychographics Subjective reality NIMCPR Effects Enculturation Social reality Norm internalization Media use patterns 4 9 --. .- .--. -- - - - - - - lo Feedback Loop 1 1 I i cultivation analysis in terms of ISP 11 ISP development & refinement 34 28ble 1 Principal Component Analysis of Youths' PTV Viewing Behaviors (with an oblique rotation of N = 254) ~~~ ~ Exposure to PTV Factor 1 Factor 2 Factor 3 hz PrOgrLUllS (religious) (dom.-ent.) (cosm.-ent.) Fahm al Qur'an .90 .19 .16 .81 Tathim al Din .81 .15 - .04 .74 Religious prog. .80 -.11 -.11 .69 Iqr2 .62 -.18 -.16 .64 Urdu-language dramas Music prog. Nelam Ghar TV-86 - .03 .03 - .09 .27 .77 - .09 .72 .04 .67 - .07 .45 - .05 .51 .68 .48 .52 Highway to Heaven -.12 - .04 - .88 .51 TV-magazine .19 .02 - .61 .47 Satdin .18 .04 - .58 .45 Knightrider - .04 .29 - .58 .35 Factor Eigenvalue 4.07 1.71 1.05 Percent variance 33.90 14.30 8.80 Table 2 Univariate F-tests of All Variables by Viewer-types Group Means Viewers Viewers Viewers Variables Relig. TV Cosm. Ent. Lhm. Ent. F P Newspaper-reported national affairs 3.40 3.11 3.08 7.85 <.001 Intrafamily interaction 3.06 2.97 2.90 1.04 ns Extrafamily interaction 3.34 2.94 3.19 3.92 <.03 Income 3.41 2.53 3.68 3.37 c.04 Age 21.29 20.06 21.07 1.77 ns Sex 1.52 1.58 1.53 .24 ns Radio news exposure 2.93 2.69 2.45 8.45 <.001 Radio national aff. interest 3.35 3.06 3.02 5.99 <.003 VCR Urdu-language movies and songs 2.76 2.94 2.93 .78 ns VCR English-language movies 1.92 2.10 1.90 .49 ns Shari'ah issue salience 3.62 3.34 3.34 3.39 <.04 Energy issue salience 3.43 3.71 3.35 3.65 c.03 Economic issue salience 3.51 3.76 3.58 2.51 ns Defense issue salience 3.79 3.85 3.71 1.71 ns Education 3.74 3.51 3.79 .36 ns Khan: Youth Viewers of Pakistan Television Table 3 Summary of Discriminant Function Analysis of PTV Youth Viewers 35 Predictors Function 1 2 Pooled within Group Correlations for Each Predictor Intrafamily interaction Shari'ah salience Extrafamily interaction Interest in radio-reported national aff. Education Exposure to radio-reported news Income Economic issue salience Newspaper-reported national affairs Energy issue salience VCR English-language movies Defense issue salience VCR Urdu-language movies and songs Sex Age Standardized Canonical Discriminant Function Coeficient Intrafamily interaction Shari'ah salience Exposure to radio-reported news Extrafamily interaction Newspaper-reported national affairs Economic issue salience .20 -.07 .24 .2 1 - .20 .28 - .22 - .07 -.lo - .04 - .04 .01 .68 .63 .4 1 .33 .28 - .20 .05 .13 - .28 .13 .01 -.18 -.05 .14 .19 .56 - .42 .27 .26 Continued oti next page ine American ~ o u r n i or isiarnic ~ ~ ~ i i l l XICILGCS y.1 Table 3 (Continued) Energy issue salience - .31 .19 Defense issue salience -.34 .oo VCR Urdu-language movies and songs .10 - .40 Income .03 - .39 VCR English-language movies - .05 .53 Eigenvalue Variance Canonical correlation Significance .21 .19 52.33 47.67 .42 .40 p=.0005 p= .W72 Note: Function 1 separates religious TV viewers (+) from domestic entertainment viewers (-1, and function 2 separates the cosmopolitan entertainment group (+) from the domestic entertainment group (-). Table 4 Canonical Discriminant Function Centroids Viewer Types Function 1 2 Religious TV viewers + .71 -.15 Domestic entertainment viewers - .33 -.19 Cosmopolitan entertainment viewers - .08 + 1.06 Table 5 Rotated Standardized Discriminant Function Coefficients: PTV Viewer-types by Demographic Categories (painvise comparisons) Religious TV Viewers vs. Domestic-Ent. Viewers Groups Predictors Function: 4 * Lo-income religious TV economic salience Hi-income religious TV age Lo-income dom. -ent . - Hi-income dom.-ent.+ intra-family interact. energy salience -.61 .60 - .45 .49 Eigenvalue .19 Canonical correlation .40 Significance I)< .005 Adolescent relig.TV(+) sex .70 Adolescent dom.-en.-(+) radio news .91 Young adult dom. ent . intra-family interact. .41 VCR Indian movies .27 Shari'ah salience .37 Young adult relig. TV- education -.66 VCR English movies - .28 Eigenvalue 3.16 .12 Canonical correlation .87 .33 Significance p<.OoO1 <.OE Religious TV Viewers vs. Cosmpolitan-Ent. Viewers 3roups Predictors Functions Lo-income religious TV age 1.16 Lo-income cosmop.-ent. economic salience - .67 Shari'ah salience .23 defense salience - .23 Eigenvalue .26 Canonical correlation .45 Significance p< .002 Adolescent relig.TV(+) education .75 Young adult relig. TV Sex - .52 1 2 Hi-income religious TV+ sex .80 Hi-income cosmop.-ent. intra-family interact. .39 Adolescent cosmop.-ent. defense salience .63 Young adult Shari'ah salience .61 intra-family interact. .42 cosmo.-ent. +(-I Eigenvalue 3.20 .14 Canonical correlation .87 .35 Significance p<.oO01 <.01 Continued on next page Table 5 (Continued) Note: - , + groups are separated on function 1; and (-) (+) groups are Religious TV Viewers vs. Domestic-Ent. Viewers Groups Predictors Functions - separate on function 2. Male relig.TV+ age .59 Female relig.TV education .36 Male dorn. -ent . economic salience - .30 Female dom . e n t . - newspap. national aff. .28 defense salience .26 intra-family interact. -.34 Eigenvalue 2.32 Canonical correlation .84 Significance p<.OoO1 Undergrad. relig.TV(+) age .67 Graduate relig.TV+ sex - .58 Undergrad. dom.-ent radio news .79 Graduate dom.-ent.(-) Shari'ah salience .59 intra-family interact. .49 economic salience .20 -.28 Eigenvalue 2.69 .12 Canonical correlation .85 .33 Significance p<.0o01 c.03 Religious TV Viewers vs. Cosmpolitan-Ent. Viewers 3roups Predictors Functions 1 2 Male religious TV+ Female religious TV Male cosmop.-ent. Female cosmop.-ent. - Eigenvalue Canonical correlation Significance age .80 energy salience .43 extra-family interac. .22 economic salience - .40 Shari'ah salience .26 2.41 .84 p < .OOol Undergrad. relig.TV(+) age Sraduate relig.TV+ sex Undergrad. cosmop.-ent Shari'ah salience Sraduate cosmop.-ent.(-) defense salience economic salience intra-family interact. Eigenvalue Canonical correlation Significance .86 -.21 .69 .24 .52 - .49 - .47 .34 2.75 .18 .86 .39 pc.Oo01 <.004 W m Khan: Youth Viewers of Pakistan 'Mevision Table 6 Prediction Results for Discriminant Solution 39 Na , Predicted Group Membership Actual Viewer Group Type Group 1 Group 2 Group 3 1 Religious TV viewers 43 G . l % 14.0% 20.9% 2 Cosmopo. ent. viewers 21 19.0% 66.7% 14.3%. 3 Domestic ent. viewers 91 30.8% 24.2% 45.1% Percent correctly classified: 53.6% ~ a150 cases were excluded from the analysis due to missing values on some of the discriminating variables.