Volume index - Journal index - Article index - Map ---- Back

Comunicar Journal 37: The University Network and on the Net (Vol. 19 - 2011)

Values perceived in television by adolescents in different cross-cultural contexts


Concepción Medrano-Samaniego

Ana Aierbe-Barandiaran

Juan-Ignacio Martínez-de-Morentín


This study was carried out on a sample group of 1,238 adolescents from eight different cultural contexts, and aimed to determine the values perceived by subjects in their favorite television characters. It also aimed to identify any possible differences between cultural contexts. The basic hypothesis for the study was that television conveys values and constitutes one of the forces for socialization at play during adolescence. The total sample group was made up by: three Spanish sub-groups, four Latin American sub-groups and one Irish sub-group. The instrument used for exploring perceived values was the Val.Tv 0.2, which is an adaptation of Schwartz’s scale. The data were collected both by means of an on-line platform and in person. In relation to the results, in general, the values most commonly perceived by adolescents are self-management and benevolence. As regards contextual differences, although significant differences were observed in all values, in no case were they particularly notable. The only exceptions were hedonism and achievement, for which no significant differences were found at all between the different contexts. The most relevant differences were found in the values of conformity, tradition, benevolence and universalism. From an educational perspective, we can conclude that the measurement instrument used may constitute an adequate tool for decoding the values perceived by adolescents in their favorite television characters.


Adolescence, cross-cultural study, favorite character, television, perceived values, socialization, measurement

PDF file in Spanish

PDF file in English

1. Introduction

We currently have enough data to enable us to state that adolescents watch television for a wide range of different reasons, such as: to entertain themselves, to learn about life or to identify with their own concerns and interests. In other words, they identify with the contents of television programs, and in turn, as with any other development context, the contents seen on the screen have both a positive and negative effect on adolescent viewers. In general, the media contribute to the development of adolescents’ identity in a reciprocal and multi-directional manner, and teenagers learn different values through media content (Aierbe, Medrano & Martínez de Morentin, 2010; Castells, 2009; Fisherkeller, 1997; Pindado, 2006).

We should not forget that, as a recent study indicates (Fundación Antena 3, 2010), adolescents spend an average of 4 hours a day engaging with different screen-based media, with television accounting for the majority of that time, despite not always being watched on a traditional TV set; and this screen time, as we may call it, affects and influences adolescents’ socialization and value acquisition process (Asamen, Ellis & Berry, 2008).

The analysis of the values conveyed through television has been the object of empirical research for over three decades now. Nevertheless, these values have changed over time, not only internationally, but within our own context also, as demonstrated by a number of recent studies (Bauman, 2003; Bryant & Vorderer, 2006; Del Moral & Villalustre, 2006; Del Río, Álvarez & del Río, 2004; Murray & Muray 2008). In general, a review of previous papers reveals a tendency to find fewer prosocial values and more materialistic values in television contents (Dates, Fears & Stedman, 2008; Méndiz, 2005), although other studies also highlight, for example, the fact that American television tends to convey altruistic behaviors (Smith, Smith, Pieper, Yoo, Ferris, Downs & Bowden, 2006).

However, a more detailed review of published findings regarding the values conveyed by television reveals that the results are disparate and complex. For example, Potter (1990) and Tan, Nelson, Dong & Tan (1997) found that television conveys the conventional values of the American middle class, and other authors such as Raffa (1983) calculated that antisocial values appeared more intensely than positive ones in television contents. However, the results found by other authors (Pasquier, 1996) indicate that television transmits both positive values and negative ones (or counter-values). In this sense, Pinilla, Muñoz, Medina & Acosta (2003) found that Colombian adolescents perceived mainly anti-values such as: jealousy, intrigue, hypocrisy and lack of respect.

Nevertheless, and this is relevant to our study, empirical evidence exists (derived from our previous work) to suggest that television viewers select the contents they watch in accordance with their own values (Medrano, Aierbe & Orejudo, 2010; Medrano & Cortés, 2007).

However, the assessment of the perception of values is a complex undertaking. In this study, we refer to the values perceived by adolescents in their favorite character, not to the values conveyed by the medium itself. In this sense, values are much more difficult to measure than other aspects of development, and moreover, the values of adolescents are particularly unstable, since these individuals are still at a stage in which their system of beliefs has yet to be consolidated. Adolescents exist in a neutral, interim phase, in which they no longer adhere to the beliefs of their previous stage, but have not yet adopted their adult views. This explains the vulnerability of the adolescent phase, and why it is so difficult not only to identify the value perceptions related to teenagers’ favorite characters, but also to work on these perceptions, due to the moment of transition through which subjects are passing. Adolescents have, without a doubt, become an important group to be considered in relation to the media (EDEDA, 2010). They are a strong presence and account for a large percentage of prime time viewers. Consequently, the offer has been expanded and adjusted to suit their tastes, among which we could highlight topics such as love, adventure, paranormal activity and music (Guarinos, 2009).

This study aims to explore the values perceived by adolescents in their favorite character, within the framework of reception theory (Orozco, 2010). Prior research into the perception of values is scarce, which is why this cross-cultural study is, in some ways, a pioneering piece of work in this field.

Despite the empirical difficulties of exploring the perception of values in television, we have based our work on the model developed by Schwartz & Boehnke (2003) and on the 10 dimensions established by these authors for their value scale, both in the 21 item and the 41 item versions. These 10 dimensions can be grouped into four large dimensions. Although this structure has a good degree of conceptual consistency, it also poses verification problems when the usual dimensional reduction techniques (Exploratory or Confirmatory Factorial Analyses) are applied. The technique used by Schwartz himself to demonstrate the consistency of his model is that of multidimensional analysis, which is a purely spatial solution to the circular configuration of this structure.

It is based on the existence of universal aspects of human psychology and interaction systems, which results in some of these compatibilities and conflicts between value types being present in all cultures, thus constituting the articulatory backbones of human value systems. In relation to the theoretical applicability of the model to different cultures, the authors highlight the existence of values which prevail not just in Spanish society, but in different cultures and countries also, such as Germany, Australia, the United States, Finland, Hong Kong and Israel. The differences between different cultures lie in the fact that some attach more importance to individualism, while others tend to prioritize collectivism. Thus, Schwartz’s values provide an empirical and conceptual framework for working in and comparing different cultures (Schwartz, Sagiv & Boehnke, 2000).

From an educational perspective, since our aim is to explore the values perceived by adolescents in television, and moreover, to do so with a cross-cultural sample group, we believe that the model proposed by Schwartz constitutes a valid, rigorous tool for the research being undertaken.

In this sense, insofar as it helps form adolescents’ personal identity, television also fosters the construction of values. The research data currently available in relation to these values are by no means homogenous. Community-type values appear least frequently during this period, although data does exist to indicate that they are in fact present, pointing to the existence of young people concerned about social justice (Jonson & Flanagan, 2000), social engagement (Bendit, 2000) and altruism as a kind of happiness akin to vertical collectivism.

Consequently, the situation in relation to current research and Schwartz’s model can be summed up as follows:

- Television, both in Spain and at an international level, provides contents which include both individualistic and collectivist values.

- Scarce empirical evidence exists regarding the values perceived by viewers in television contents.

- Adolescents tend to perceive their own values in television.

- Schwartz’s model is an adequate tool for exploring the values perceived by adolescents from different cultures in their favorite television character.

- Differences regarding values in different cultures are polarized between individualism and collectivism.

In accordance with the prior review of the literature, this study has a twofold objective:

- to identify the values perceived by a cross-cultural sample group of adolescents in their favorite character, within the framework of Schwartz’s model; and

- to analyze similarities and differences in the values perceived by adolescents in different cultural contexts.

2. Materials and method

This research project is an ex post-facto, descriptive-correlational, cross-cultural study. It explores the values perceived in the characters of the favorite TV programs of a group of adolescents aged between 14 and 19. 44.6% of the sample group were male and 55.4% female.

2.1. Participants

Once all the extreme cases (such as subjects who gave inconsistent responses) had been eliminated from the different cultural contexts, the sample group was distributed as follows:

The total sample group encompassed 1,238 subjects from 8 different cultural contexts; three from Spain, four from Latin America and one from Ireland.

The gender percentage was balanced for all cities. Nevertheless, in San Francisco de Macorís and Rancagua, the percentage of male subjects was 28.1% and 35% respectively. In total, the sample group was comprised by 545 boys and 676 girls. Some cases were lost, since no information was provided regarding this variable.

The sample group was selected on the basis of convenience, in accordance with the following criteria: age, academic year and type of school. Subjects were from the 4th year of secondary school or the 2nd year of the Spanish Baccalaureate (higher education) system, equivalent to the Latin American PREPA and/or Baccalaureate years 1 and 3 and the 3rd year of Junior Certificate and 2nd year of Leaving Certificate in Ireland. As regards type of school or college, the sample group was taken from two or more schools for each sub-sample (city), both state and private, or with different socioeconomic levels (although no extreme cases were included).

The 23 schools and colleges from which the sample group was selected were distributed as follows: Málaga (2 schools, one private and the other state); San Sebastián (2 schools, one state and the other private but with some state funding); Zaragoza (2 schools, one state and the other private); Rancagua (Chile) (2 schools, one state and the other private); Guadalajara (Mexico), (one private middle-class school); Macorís (Dominican Republic) (2 schools, one state and the other private); Oruro (10 state and private schools) and Dublin (2 schools, one state and the other private).

2.2. Variables and measurement instruments

The instrument used to assess the values perceived in the character of subjects’ favorite TV show was the Spanish language version of the PVQ-21 scale by Schwartz (2003), called Val.Tv 0.2. The scale measures the values perceived in subjects’ favorite characters, grouped into 10 basic values: self-direction, stimulation, hedonism, achievement, power, security, conformity, tradition, benevolence and universalism. The scale consists of 21 items, the responses to which are scored on a Likert scale, from one to six.

In order to check the reliability of the instrument, the spatial configuration of these items was verified using the “multidimensional scaling” technique presented by the SPSS, which is very similar to the SSA (Small Space Analysis). The configuration obtained was circular, very similar to that proposed by Schwartz and Boehnke (2003). Nevertheless, one exception should be highlighted: the value “power” was found to have an inadequate spatial configuration.

This spatial configuration, which is similar to that proposed by Schwartz, coupled with the fact that this instrument is comparable to those used in international studies, enables us to calculate the scores for each subscale or dimension. The analysis of the internal consistency of each dimension using Cronbach’s alpha coefficient resulted in reliability indexes of over 0.50 (the maximum was obtained for universalism a=.798 and the minimum for security a=.529), with the exception of the value “power”, which had an a =.389.

2.3. Procedure

For the data collection process, the first phase consisted of adapting the Val.Tv 0.2 scale from the Spanish version to a Bolivian, Chilean and Mexican version. Also, the original version was translated and adapted to an English version. These adaptations were all carried out without changing the meaning of the values. The scale was revised by eight experts prior to its definitive approval. In addition to other aspects, the experts were asked to assess whether the definitions of the values were applicable to each culture. The majority of participants completed the questionnaires on-line. In the Bolivian and Dominican samples, the data were collected on paper, due to the lack of computer facilities required for this kind of on-line questionnaire. Subsequently, the data gathered on paper were entered into an on-line version, in preparation for their statistical processing.

Between 20 and 30 minutes are required for the application of this scale. In relation to the data analysis, the SPSS program was used and a number of different descriptive and inferential analyses were carried out, mainly the means comparison test and parametric tests (Anova). These not only enabled the identification of the different means, but also verified the significance of these results and the effect size, or in other words, the magnitude of said differences.

3. Results

The analysis of the results focused on the study’s two established objectives. Thus, first of all, the means of the values perceived in subjects’ favorite characters were analyzed for the whole sample group, along with their significance level. This was then followed by an analysis of the differences between the eight sub-sample groups studied in relation to the values perceived in each one.

If we focus on the first objective, as shown in Table 2, only some items presented asymmetries higher than 1, and almost none of them were over 1.5. It was therefore considered that these variables were susceptible to processing using parametric tests.

In the first analysis, as shown in Table 2, the highest means were obtained for self-direction (items 1 and 11), with means of 5.03 and 5.01 respectively, followed by benevolence (items 12 and 18), with means of 4.70 and 4.79 respectively. On the other hand, the values with the lowest means were item 2 (power), with a mean of 3.14 and items 7 and 16 (conformity), with means of 3.41 and 3.96, respectively. We believe it is important to highlight the fact that self-direction can be understood as an individualistic value, while benevolence is conceptualized as a collectivist value.

Thus, self-direction is defined as applying to “an active person with independent thought. Also a person who stands out for their creativity, freedom and for choosing their own goals. The value ‘benevolence’, on the other hand, is understood as applying to a person who values the preservation and enhancement of the welfare of people with whom they have frequent personal context; a benevolence person is helpful, honest, indulgent, loyal and responsible.”

Thus, the two values with the highest scores effectively reflect two different dimensions which are not necessarily contradictory, as explained in the conclusions.

However, both the value “power”, which is defined as applying to a person who values social status and prestige, and who seeks to gain control or dominance over people and recourses, and the value “conformity”, which is defined as applying to a person characterized by their restraint of those actions, inclinations and impulses likely to upset or harm others and violate social expectations or norms, obtained the lowest means. The value “power” is also an individualistic value, while “conformity” is collectivist and refers to a person who is excessively conventional or concerned about living up to others’ expectations.

In relation to the second objective, as shown in Table 3, following an analysis of the differences between contexts, in general terms we can state that significant differences were found between the eight cultural contexts studied. Nevertheless, it is important to remember that the value “power” was not included in the calculations, due to its low level of reliability.

On an initial reading, clear differences emerge between the different cities; thus, San Sebastián, followed by Zaragoza, have the highest means (5.097 and 5.003 respectively) in the self-direction dimension.

If we look at benevolence (items 12 and 18), another of the values with the highest scores, San Francisco de Macorís has a mean of 5.162, followed by Oruro with a mean of 4.875. These scores are the highest of all those obtained on a scale where the maximum score was six. We should remember that when we talk about benevolence, we are talking about people who value the preservation and enhancement of the welfare of the people with whom they have frequent personal contact.

However, the lowest values were found for conformity (items 7 and 16). Thus, the mean for Dublin was 3.332 and the mean for Zaragoza 3.385. Conformity is understood as referring to a person characterized by their restraint of those actions, inclinations and impulses likely to upset others. It was important to determine the statistical significance of these differences, and their magnitude. The Anova revealed that all differences were statistically significant (p=.000) except in the dimensions of hedonism F (7.1165=.753; p=.627) and self-promotion F (7.1158=1.834; p=.077).

Nevertheless, when the effect size was analyzed taking these differences globally, they were found not to be great enough to be notable. Having conducted the “eta” coefficient analyses, it can be stated that the differences between the eight contexts studied are hardly relevant at all. The highest results were found for universalism (?2=.073), followed by tradition (?2=.068) and conformity (?2=.042), and the lowest results were found for hedonism (?2=.005), followed by stimulation (?2=.021).

In other words, the greatest differences were found in items 7, 16, 9, 20, 12, 3 and 8, which belong to the values of conformity (which scored very low in all contexts), tradition (with a mean of 3.804 in Dublin and 4.910 in San Francisco de Macorís) and benevolence (with a mean of 4.688 in Málaga and 5.162 in San Francisco de Macorís). In relation to universalism, the mean for Dublin was 3.592, and the mean for San Francisco de Macorís was 4.983.

4. Conclusions

If we carry out a global assessment, taking all the data into account and focusing on the pre-established objectives of the study, we can state that adolescents perceive both individualistic values (self-direction) and collectivist values (benevolence) in their favorite TV character, while the values “power” and “conformity” seem to be perceived less by the sample group on the whole. Cross-cultural differences were found in all values, with the exception of hedonism and achievement. However, these statistically significant differences are not particularly relevant, as shown by the analyses of the results. It is likely that this is due to the size of the sample, since in a detailed analysis of the data, these mean differences are not particularly extreme, and no clear differentiating trends can be detected between the different cultures. In other words, our data do not enable us to state that some cultural contexts are more inclined to perceive individualistic values, while others tend more to perceive collectivist ones.

These results coincide with those of the studies cited in the introduction to this paper (Bendit, 2000; Dates, Fears & Stedman, 2008; Del Río, Álvarez & Del Río, 2004; Méndiz, 2005). As in previous research projects, our results also indicate that television conveys individualistic or presentist values, while at the same time transmitting prosocial values. Nevertheless, it is important to clarify that our data refer only to the values perceived by adolescents.

Prior research concludes that television conveys both individualistic and collectivist values, values which are not only typical of adolescents, but which also constitute one of the characteristics of postmodern society (Castells, 2009; Goldsmith, 2010; Maeso Rubio, 2008). In our research, however, although we define the value self-direction as the achievement of personal goals, we should not necessarily interpret this to imply that self-direction clashes with benevolence. Both values, self-direction and benevolence, enable a person to be competitive and to develop their abilities to the full (achievement), while at the same time being concerned about others. These tendencies are perfectly compatible from the perspective of value development.

Nevertheless, one aspect of our data that does warrant attention is the absence of major contextual differences; this may indicate a globalization of the values conveyed and perceived through television. Although this claim requires further exploration through more qualitative studies using semi-structured interviews with young people from different cultures, the working hypothesis is of enormous interest in our attempt to understand how television may foster socialization and the acquisition of values during adolescence, a phase of particular vulnerability. Could it be that in the global village, the perception of values by adolescents is becoming increasingly similar? An initial reading of our data indicates that this is indeed the case, and moreover, the value “power” and the value “conformity” are those perceived less frequently. Bearing in mind that a relationship exists between our values and the values perceived through television, this would seem to indicate that the adolescents in our study evince little interest in becoming successful, ambitious and/or influential people (power). Furthermore, they perceive those characters they “like best” as fairly unconventional. Bearing in mind the age range of the sample group and that one of the characteristics of adolescence is the transgression of conventional norms, this last result is consistent with the assessment instrument itself.

Do these findings indicate a trend in the values of today’s society? Many authors have already warned of the ambivalence of value transmission; however, the adolescents in our study did not perceive ambivalent values and curiously enough, one of the quintessential individualistic values, power, was found to be one of the least perceived. The fact that values such as conformity, defined as restraint of those actions, inclinations and impulses likely to upset or harm others and violate social expectations or norms, as well as courtesy, obedience, self-discipline and honoring one’s parents and the elderly, scored lowest in our study is extremely interesting, and to some extent contradicts the findings of other research projects. These data are not consistent with the results of other research studies carried out in the American context, which assert that television conveys the conventional values of the middle class (Potter 1990; Tan, Nelson, Dong & Tan 1997).

Before concluding, we would like to highlight some limitations of the study, and to add a comment which goes beyond the scope of the specific results. As regards the limitations, might it not be that the design of the trial itself (a six-point Likert scale with the definition of the values in each item) fosters the social desirability bias? In other words, could it be that respondents give what they consider to be politically correct answers, rather than state what they really feel? Moreover, we should not forget that, in accordance with the reception theory (Orozco, 2010), adolescents do not perceive messages as blank pages, open to any manipulation, but rather interpret them in accordance with their own prejudices, values and ways of thinking, etc.

As regards the comment, we should point out that, in addition to its assessment function, the Scale of Television Value Domains (Val.Tv 0.2) can also be used as an instrument for fostering the explicit expression of the values perceived by adolescents in the characters of their favorite TV shows. Prompting adolescents to explicitly identify and reflect upon values is an important strategy within the psychoeducational field. The instrument presented here may help us translate the implicit messages conveyed by television, share them with others and develop a critical attitude to them. As shown by a number of different authors (Lejarza, 2010; Rajadell, Pujol & Violant, 2005), television may have an enormously constructive effect on the dissemination of values which render learning attractive and which promote the need to make an effort to acquire knowledge. There can be no doubt that teaching about and sharing television is much more effective than restricting or limiting it.


This piece of research forms part of a project funded by the Ministry of Education and Science, reference num. EDU2008-00207/EDUC, within the framework of the National Scientific Research and Technological Development and Innovation Plan.


Aierbe, A.; Medrano, C. & Martínez de Morentin, J.I. (2010). La privacidad en programas televi-sivos: percepción de los adolescentes. Comunicar, 35; 95-103.

Asamen, J.H.; Ellis, M.L. & Berry, G.L. (2008). Child Development, Multiculturalism, and Media. Londres: Sage.

Bauman, Z. (2003). Comunidad. En busca de seguridad en un mundo hostil. Madrid: Siglo XXI.

Bendit, R. (2000). Adolescencia y participación: una visión panorámica en los países de la Unión Europea. Anuario de Psicología, 31, 2; 33-57.

Bryant, J. & Vorderer, P. (2006). Psychology of Entertainment. Mahwah, NJ: Lawerence Erlbaum Associates.

Castells, M. (2009). Comunicación y poder. Madrid: Alianza.

Dates, J.L.; Fears, l.M. & Stedman, J. (2008). An Evaluation of Effects of Movies on Adolescent Viewers, en Asamen, J.H.; Ellis, M.L & Berry, G.L. (Eds.) Child Development, Multiculturalism, and Media. London: Sage; 261-277.

Del Moral, M.E. & Villalustre, L. (2006). Valores televisivos versus valores educativos: modelos cuestionables para el aprendizaje social. Comunicación y Pedagogía: Nuevas tecnologías y recursos didácticos, 214; 35-40.

Del Río, P.; Álvarez, A. & Del Río, M. (2004). Pigmalión. Informe sobre el impacto de la televisión en la infancia. Madrid: Fundación Infancia y Aprendizaje.

Entidad de Gestión de los Derechos de los Productores Audiovisuales EGEDA (2010). Panorama audiovisual, 2010. Madrid: EGEDA.

Fisherkeller, J. (1997). Everyday Learning about Identities Among Young Adolescents in Television Culture. Anthropology and Education Quaterly, 28, 4; 467-492.

Fundación Antena 3 (2010). En busca del éxito educativo: realidades y soluciones. Madrid: Fun-daciones Antena 3.

Goldsmith, S. (2010). The Power of Social Innovation. San Francisco: Jossey Bass.

Guarinos, V. (2009). Fenómenos televisivos ‘teenagers’: Prototipias adolescentes en series vistas en España. Comunicar, 33; 203-211.

Jonson, B. & Flanagan, C. (2000). Les idées des jeunes sur la justice distributive, les droits et les obligations: étude transculturelle. Revue International des Sciencies Sociales, 164; 223-236.

Lejarza, M. (2010). Desde los medios de comunicación: Enseñar y compartir la televisión, en Maeso Rubio, F. (Ed.). La TV y la educación en valores. Comunicar 31; 417-437.

Medrano, C. & Cortés, A. (2007). Teaching and Learning of Values through Television. Review International of Education, 53, 1; 5-21.

Medrano, C., Aierbe, A. & Orejudo, S. (2010). Television Viewing Profile and Values: Implications for Moral Education. Revista de Psicodidáctica, 15, 1; 57-76.

Méndiz, A. (2005). La juventud en la publicidad. Revista de Estudios de Juventud, 68; 104-115.

Murray, J.P. & Murray, A.D. (2008). Television: Uses and Effects, en Haith, M.M. & Benson, J. B. (Eds.). Encyclopedia of Infant and Early Childhood Development. NY: Academic Press; 309-318.

Orozco, G. (2010). Niños, maestros y pantallas. Televisión, audiencias y educación. Guadalajara (México): Ediciones de la Noche.

Pasquier, D. (1996). Teen Series Reception: Television, Adolescent and Culture of Feelings. Child-hood: A Global Journal of Child Research, 3, 3; 351-373.

Pindado, J. (2006). Los medios de comunicación y la construcción de la identidad adolescente. Zez, 21; 11-22.

Pinilla, E.R.; Muñoz, L.R. & al. (2003). Reality Show en Colombia: sobre la construcción social de valores. Bogotá: Comisión Nacional de Televisión CNTV.

Potter, W.J. (1990). Adolescents´perceptions of the Primary Values of Televisión Programming. Journalism Quarterly, 67; 843-853.

Raffa, J.B. (1983). Televisión: The Newest Moral Educator? Phi Delta Kappan, 65, 3; 214-215.

Rajadell, N.; Pujol, M.A. & Violant, V. (2005). Los dibujos animados como recurso de transmisión de valores educativos y culturales. Comunicar, 25.

Schwartz, S.H. & Boehnke, K. (2003). Evaluating the Structure of Human Values with Confirma-tory Factor Analysis. Journal of Research in Personality, 38; 230-255.

Schwartz, S.H.; Sagiv, L. & Boehnke, K. (2000). Worries and Values. Journal of Personality, 68; 309-346.

Smith, S.W.; Smith, S.L. & al. (2006). Altruism on American Television: Examining the Amount of, and Context Surronding. Acts of Helping and Sharing. Journal of Comunication, 56, 4; 707-727.

Tan, A.; Nelson, L.; Dong, Q. & Tan, G. (1997). Value Acceptance in Adolescent Socialization: A Test of a Cognitive-functional Theory of Television Effects. Communication Monographs, 64; 82-97.