publisher colophon
  • Does Media Literacy Work? An Empirical Study of Learning How to Analyze Advertisements

Many school districts are implementing media literacy programs in high schools that teach about the advertising production process and introduce students to techniques for critically analyzing media messages. In this study, students who learned how to critically analyze advertising as part of their Grade 11 English language arts class were compared to a demographically matched control group who did not receive such instruction. Four weeks of classroom activities involving the analysis of the purpose, target audience, point of view, and persuasive techniques used in advertising were provided as a regular part of classroom instruction in English language arts to 293 students by seven regular classroom teachers. Compared to the control group, students gained increased knowledge of the pre-production processes of advertising. Statistically significant differences were also found in measures of students’ ability to analyze a print ad, including the ability to identify target audience, to describe construction techniques used to attract and hold attention, and the ability to identify the implied message subtext.

Advertising is increasing its prominence in American public schools, but not just as components of particular marketing campaigns. In an increasing number of secondary classrooms, print and TV ads are used by teachers as texts to be formally analyzed and studied. Educational practices like this are commonly identified as media literacy, which is defined as an expanded conceptualization of literacy that includes print, audio, visual, and electronic messages from contemporary culture (Kress, 2002). In using advertising texts in the classroom, teachers emphasize the skills of analyzing and evaluating ads to identify the message purpose, target audience, point of view, and persuasive techniques used. Often, there is a focus on the social, political, economic, and historical contexts in which media messages reflect and shape culture (Buckingham, 2003).

Occasionally, as part of media literacy education, students also learn about the preproduction, production, and postproduction processes involved in the creation of advertising messages (Young, 1990; Singer, Zuckerman, & Singer, 1980). While it may be common for students enrolled in media production or marketing electives to learn about advertising production processes, it is far less common for students to gain this information in the context of their high school English coursework. Potter (1998) points out the importance of knowledge structures in building critical analysis skills when it comes to analyzing advertising, but empirical research has not yet examined the impact of increased knowledge of advertising production processes as it may affect critical thinking skills in responding to advertising messages.

Drawing upon a tradition underway for the last 15 years (see Alvarado & Boyd-Barrett, 1992 and Brown, 1991 for reviews), in the United Kingdom, Canada, and Australia, advertising messages are routinely included in the study of persuasion and propaganda, and students are invited to design and create advertising in print, visual, video, and multimedia formats as part of their instruction in understanding concepts like target audience, purpose, point of view, persuasive techniques, and message interpretation and impact.

In the United States, a coalition of U.S. educators with interests in helping children and young people strengthen their media analysis skills has formed a national association and held bi-annual conferences bringing K-12 educators together with academics and community activists (Rogow, 2001). There has been increased momentum to include media literacy skills within state curriculum frameworks, so much so that secondary level English language arts textbooks now include the formal study of advertising (Odell, Vacca, Hobbs, Irvin & Warriner, 2000). In 1998, the State of Texas added viewing and representing skills alongside reading, writing, speaking and listening skills for students in Grades 4–12, and state curriculum frameworks make explicit reference to the genres of advertising, documentary, film drama, and news (Texas Education Agency, 1998).

The ability to analyze advertising was recently included as a component of the high-stakes TAAS test for 10th graders in Texas, involving students’ ability to identify the persuasive techniques used in a specific print ad (Livaudais, 2002). More than 40 states including Massachusetts, North Carolina, and New Mexico have identified media literacy skills within language arts, social studies, fine and performing arts, library information skills, or health education curricula (Kubey & Baker, 1999).

There is a small body of research that explores the impact of media literacy instruction on the cognitive skills, attitudes, and behaviors of young people. A history of the first phase of implementing critical viewing skills instruction in the 1980s revealed that most evaluation models examined the program outcomes on very small numbers of students, usually a single classroom, often in interventions designed and implemented by researchers (Anderson, 1983). Studies have examined whether a brief, six-hour exposure to media literacy education affected children’s ability to distinguish between realistic elements (like location or setting) and fictional elements (like characters and plot) of a program (Dorr, Graves & Phelps, 1980); whether a three-hour-a-week curriculum for elementary school students helped students identify genre and syntactical structure (Anderson, 1983); and whether an eight-session course on media literacy improved knowledge of camera and editing production techniques and the economics of media production (Singer, Zuckerman & Singer, 1980). More recently, studies have explored whether students learned the facts, vocabulary, and information provided as part of the instruction (Baron, 1985) or whether a video broadcast about media literacy affected cognitive or critical analysis skills (Vooijs & Van der Voort, 1993).

Health researchers have examined the effect of media literacy instruction on elementary school students’ attitudes about alcohol (Austin & Johnson, 1997; Goldberg & Bechtel, no date). School-based empirical research has also been conducted to demonstrate the impact of media literacy curricula on students’ attitudes, behavior, knowledge, and academic performance. Quin and McMahon (1995) conducted research on a sample of 1500 students in Western Australia. They created an evaluation instrument that provided students with a specific visual media message, using multiple-choice and open-ended questions in a paper-and-pencil assessment. Students identified the message’s purpose, target audience, point of view, and qualities of representation.

There is only limited evidence that shows that learning about advertising and discussions about advertising in school can reduce children’s vulnerability to advertising appeals and increase their ability to produce counterarguments in response to advertising. For example, Christenson (1982) developed a three-minute video about advertising and showed it to children ages six to 12, finding that children who viewed the video were more aware of commercials and expressed less trust in commercials in general. Roberts, Christenson, Gibson, Mooser, and Goldberg (1980) evaluated short films that were made to show children how television ads use various techniques to persuade. They found that heavy-viewing children who were initially most susceptible to commercials were most influenced by the films. In a correlational study, Slater, Rounder, Murphy, Beauvais, Van Leuven and Domenech-Rodriguez (1996) found that 12- to 18-year-old students who discussed alcohol advertising in school during the previous year provided nearly twice as many counterarguments after viewing a 20-minute sports or comedy program with ads including four beer advertisements as compared with students who did not receive an educational intervention in analyzing alcohol advertising.

Thirty years ago, Rossiter and Robertson (1974) formulated the concept of cognitive defenses, the idea that youth can become capable of resisting advertising appeals by perceiving the purpose behind the message and understanding how persuasive techniques work. These researchers postulated that understanding the selling intent of ads requires five component skills:

  1. 1. distinguishing program from commercial;

  2. 2. recognizing that an ad was created by an author or external source;

  3. 3. appreciating that an ad is designed to reach an intended target audience;

  4. 4. identifying the symbolic and constructed nature of advertising; and

  5. 5. being able to generate real-world examples of a product not meeting the expectations generated by the advertising.

Since the publication of this article, there has developed a considerable, if disparate, literature measuring young children’s understanding of advertising, although there is considerably less evidence available concerning older children and teens. Researchers have examined children’s ability to identify the purpose of advertising, finding that by age ten or 11, children are able to identify the persuasive intent of advertising, with younger children seeing commercials as largely informative (see Young 1990 for review). Ross, Campbell, Huston-Stein, and Wright (1981) found that there were no differences between sixth graders and younger students in their susceptibility to strategies like celebrity endorsement, finding that children from ages five to 12 were effectively persuaded by advertising’s common tactics in appealing to youth, and did not activate critical thinking skills or counterarguments.

Researchers have also examined children’s levels of skepticism towards advertising. Brucks, Armstrong, and Goldberg (1988) measured fourth graders’ responses to advertising using an open-ended cognitive response measurement, a technique developed by Petty, Cacioppo, and their associates in which respondents view or read a message and then list the thoughts and feelings that they recall arising during the processing of the message in response to one or more probes. Researchers found a relationship between knowledge about advertising and children’s ability to generate skeptical and critical ideas in response to direct questions about advertising. However, when they measured students’ responses to advertising through an open-ended (non-cued) question, children did not generate critical thoughts during ad exposure. This evidence suggests that unless children under the age of twelve are cued to respond cognitively to an ad, they may not activate their knowledge of advertising in responding to the message.

While older children and teens may have more knowledge about advertising, they also may not necessarily employ critical thinking skills in response to advertising or have more skepticism about advertising in general. Boush, Friestad, and Rose (1994) measured middle-school students’ knowledge of advertiser tactics and effects, and their skepticism of advertising. Tactics included the use of celebrities, music, humor, cartoons, product comparisons, product demonstrations, and depictions of target audience. They found increased knowledge about advertiser tactics over a nine-month period, but no increase in advertising skepticism. They found an association between knowledge about advertising tactics and being more skeptical of advertising. They note:

Improving students’ understanding of the way advertising works may have more potential for creating discerning consumers than has changing students’ general attitudes . . . Exhortations to ‘not believe everything you see on TV’ are, therefore, less likely to produce changes in the processing of advertising claims than is a more careful analysis of advertisements that lays bare the persuasive device

(p. 172).

In evaluating the literature on advertising and children, Young (1990) criticizes the validity of research that has used superficial measures of children’s skepticism including responses to attitude statements using Lipert-type scales. Even young children are aware of the social desirability of attitudes opposing advertising, he claims. Instead, Young argues that knowledge about the tactics used by advertisers to persuade and skills like being able to understand the purpose and function of a media message are key components needed to acquire critical thinking skills about advertising.

The present study was designed to evaluate the impact of a large-scale, year-long implementation of media literacy in a secondary language arts curriculum. This study sought to determine specifically whether instruction in media literacy affects students’ knowledge of advertising production processes and the ability to critically analyze alcohol advertising by identifying the presumed target audience, use of visual and verbal construction techniques, message purpose, and subtext.

A School District Integrates Media Literacy into the Curriculum

Alverson High School (pseudonym) is one of a small number of high schools in the United States to fully integrate media literacy for all its students. During the spring of 1998, the school board approved a plan to reorganize the high school English language arts curriculum to include a full year-long curriculum in communications/media for all Grade 11 students. Working collaboratively, teachers designed a year-long program in communications/media that emphasized five broad themes:

  1. 1. the analysis of media messages using five critical questions to promote critical thinking and inquiry;

  2. 2. advertising, propaganda, and the manufacture of consent;

  3. 3. the nature and function of storytelling in a culture;

  4. 4. media representation of age, gender, race, ethnicity and ideology; and

  5. 5. the structure and function of contemporary journalism and the impact of the news media on American society.

Seven different teachers collaborated to construct and implement the curriculum with over 300 students. Teachers met with students four days per week for 90 minute blocks during an entire school year. Activities during the year involved students analyzing the language and images of classic and contemporary literature as well as television shows, print and television journalism, films, advertising, political speeches, and business and interpersonal communications. Numerous reading, writing, viewing, listening, public speaking, research, and small and large group discussion experiences were part of the classroom routine.

Teachers designed specific instructional activities to focus on advertising and implemented them for four weeks. For example, students analyzed the techniques and approaches used in print and TV advertising. Many writing assignments encouraged students to examine ads and describe target audiences, recognize the use of emotional appeals, and notice how graphic design elements were used to compel viewer attention. Students visited an advertising agency and interviewed key staff members. Some taught a mini-unit on advertising to younger children, created ad parodies, or constructed consumer awareness campaigns using flyers, radio advertising, and print media. While teachers did share materials and resources with others, each teacher designed and used various units of instruction according to their individual priorities and perspectives. Because of the need to share videos, materials, and books, each teacher taught using these materials at different times and in different sequential order during the school year.

Teachers adopted five framing questions to help to unify their year-long curriculum. These questions were routinely used to analyze different types of media messages. The questions include:

  1. 1. Who is sending this message and what is the author’s purpose?

  2. 2. What techniques are used to attract and hold attention?

  3. 3. What lifestyles, values and points of view are represented in this message?

  4. 4. How might different people interpret this message differently?

  5. 5. What is omitted from this message?

Research Design

Research Hypotheses

We wanted to determine whether instruction in media literacy affects students’ ability to critically analyze advertising and sought to measure changes in students’ knowledge of the pre-production, production, and post-production processes used in television advertising. The following were stated as null hypotheses:

H1. Media literacy instruction does not increase students’ knowledge about pre-production, production, and post-production processes involved in the creation of a TV ad.

H2: Media literacy instruction does not increase students’ advertising analysis skills. Students who receive instruction in media literacy will not have stronger skills in identifying a print ad’s purpose and target audience, identifying message construction techniques, and identifying the subtext (implied or unstated message) of both the ad and a visual logo contained in the ad.

Because all students in the school district participated in the instructional program at Alverson High School, it was necessary to use a non-equivalent groups design in order to quantitatively measure the impact of this instructional treatment on student learning. While such designs are common in evaluation research (Babbie, 1998), they have been critiqued by both by advocates of single-subject research and by advocates of qualitative research, who point out the limitations of the method (Kennedy, 1997; Richardson, 1994). However, non-equivalent research designs are still a powerful tool for understanding the effectiveness and impact of new instructional interventions (Cook & Campbell, 1979; Gersten, Baker & Lloyd, 2000). In true experiments, participants are randomly assigned to treatment conditions. In quasi-experiments, researchers often use students from intact classes or schools as the treatment sample and try and find a relatively comparable group of students from other classes or schools to serve as the comparison sample. Since randomized assignment to treatment groups was impossible because the entire population of the school was implementing the media literacy curriculum, a comparison sample was used from another community with similar instructional quality, school size, and student demographics.

However, even when control and experimental groups are matched along key demographic variables and other criteria, selection bias remains a threat to internal validity, which limits the generalizability of the results. However, the primary advantage of this design is that it allows researchers to eliminate maturation effects, thus distinguishing between effects of the instruction and those of natural developmental maturation and growth (Cook & Campbell, 1979).


Data were gathered on 293 students at Alverson High School enrolled in Grade 11 and a random sample of 89 Grade 11 students from a control school, located within a 50-mile radius of the treatment school. Demographic comparison of the two communities shows a pattern of similarities in size, racial, gender, and social class variables. Both Alverson and the control community each have a population of approximately 7,000 families, 97% white. Alverson’s per capita income is slightly lower (at $23,262) than the control community (at $24,367) because Alverson has a greater number of elderly citizens. Students in both samples have a balanced proportion of male and female students, and both groups are matched in the racial composition of the communities.

A comparison of parental occupations revealed parity between the two communities. We asked students to write down parental occupations as an indirect measure of socioeconomic status and then coded these using categories identified by the U.S. Department of Commerce. For example, 27% of Alverson fathers and 23% of control group fathers are identified as book and business knowledge intensive (includes managerial, finance, legal, government); 18% of Alverson and 22% of control group fathers are identified with science and technology intensive jobs (includes computers, engineering); 24% of both Alverson and control group fathers are identified with hands-on work (includes military, construction, installation, maintenance); and 18% of Alverson and 15% of control group with work that is communication intensive (includes community and social services, education, and managerial). Because of the parallel proportions of mothers and fathers involved in hands-on work, service occupations and knowledge professions, this data enhances our confidence that the samples, while drawn from two different communities, are similar along key dimensions of race, educational background, employment, and social class.

Media use data reveal few differences in the media consumption patterns between the control group and treatment group. Media use indicators showed no significant differences between the control and the treatment group in the number of televisions, number of videocassette recorders, cable television, and newspaper subscriptions. Since a greater proportion of control group students had a computer in the home (96% as compared with 90% among treatment group, F (1, 726) = 7.51 p < .001), we used this variable as a covariate in the statistical analyses performed.

Family size data suggest that the students receiving media literacy instruction may have a slightly higher proportion of smaller, possibly single-family households. While there are fewer siblings in the treatment group (1.8 for the control group and 1.7 for the treatment group) there are small but statistically significant differences in total household size in the treatment group (control group M = 4.2, treatment group M = 4.0, F (1, 726) = 6.82 p < .01).

In addition to the close demographic match, the control school was also selected because of its similarities in terms of the overall instructional program and the quality of the program in English language arts. The principal of Alverson High School had previously served as an administrative leader at the control group school and recommended the school’s participation in the research because of perceived similarities in the quality of the faculty, parental backgrounds, funding priorities in the district, and the overall administration of the school. Like the treatment group, the English language arts faculty at the control school favored heterogeneous grouping and had an outstanding reputation in the state for excellence in achievement. The control school’s English 11 curriculum emphasized world literature, and like the Alverson teachers, the control school faculty valued a process approach to writing, emphasis on critical thinking, rich discussion, and collaborative learning.

Expectations for student learning were high in both schools, according to the instructional leaders who were interviewed. During the testing year faculty at the control site were themselves excited to be involved in pilot testing a new program of portfolio-based assessment. According to the curriculum coordinator for the school, there was significant enthusiasm for this new evaluation approach, which the faculty had long promoted. Differences in teacher enthusiasm can confound results of quasi-experimental designs (Scruggs & Mastropieri, 1994), and while this research did not formally measure teacher enthusiasm in either the treatment group or the control group, interviews with teachers from both sites gave us the indication that faculty were equivalent in their engagement with students and level of morale.

Because data were collected from the entire population of Grade 11 students at Alverson High School, the treatment sample included all students enrolled in the regular and special education programs. Because courses of study at this school were heterogeneously grouped, the sample included students with learning disabilities, physical disabilities, and hearing impaired students. Only students who completed the entire battery of identical pretest and posttest measures (administered in September as a pretest and in May as a posttest) were included in the study. Although we were unable to collect data from the entire population at the control school, we were able to use a random sample, which included students with learning disabilities, physical disabilities and hearing impaired students. Students in both the control and treatment samples were an average of 17 years old at the start of the testing, with a range of 16 to 18 years.


This study measured students’ knowledge of TV advertising production processes and advertising critical analysis skills in response to TV and print ads. The study employed a pretest/posttest, control group/experimental group design. To measure students’ knowledge of advertising production processes, students were shown a 30-second Pepsi commercial featuring Cindy Crawford, who is shown drinking a can near a roadside gas station while two boys watch in awe. After viewing, students were given a lined sheet of paper and asked to create a step-by-step list of the activities that were involved in constructing the ad.

To measure media literacy skills, students received a black-and-white copy of a print alcohol ad for Miller Beer.1 The ad featured two young black men dressed in suits, one holding a saxophone and the other holding a can of beer. The phrase, “Life in the Cold Lane” was prominent, and the lower right hand corner of the ad contained a small logo featuring a traffic yield sign with the words, “Think When you Drink.”

No description available
Click for larger view
View full resolution

Beer ad

Students were asked to complete a series of paper-and-pencil response questions while viewing the ad. Students’ analysis skills were measured using both open-ended and checklist items to determine students’ ability to identify purpose, target audience, construction techniques, and message subtext. Appendix A displays the questionnaire.

Click below to view Appendix

Identical test administration procedures and measures were used for both pretest and posttest for the treatment and control groups. The use of a written protocol ensured standardized administration procedures for the testing. Students entered a study hall in groups of 30 to 50, accompanied by classroom teachers, who stayed in the room throughout the administration of the measures. Students received code numbers on the first day of testing, and all data collected in this study kept students’ names confidential.

Decisions about scoring began after reviewing a sample of 40 student responses and reviewing the test responses supplied by the seven Grade 11 treatment group faculty who also completed all tests. Researchers created a coding protocol by first identifying a range of possible written responses to each item. The coding protocol was a detailed written guide to assist scorers in identifying how to allocate points for student responses. Two advanced undergraduate students who served as peer writing coaches at their college’s writing center read all student responses and conducted the scoring after a training process. The author served as a guide to coders’ decision making during the initial period of learning to use the scoring instrument. A random sample of tests revealed a Cronbach’s alpha of inter-rater reliability ranging from .89 to .93 for items requiring the scoring of open-ended responses. Coders read and scored the data blind to pre/post and control/treatment condition.

Measuring Knowledge of TV Advertising Production Processes

In examining students’ knowledge of TV advertising production processes, students were asked to list all the steps involved in the production of the TV ad they had viewed. The instructions stated: “Many people were involved in creating the TV ad you just viewed. These people engaged in a variety of tasks to produce the ad. On the lines below, list all the steps you can think of that were probably involved in the process of creating this ad from start to finish. Example of one step: A script is written.” After reading a sample of student responses and based on the researcher’s knowledge of advertising production, we identified eight pre-production steps, two production steps, and seven post-production steps. Students’ responses were coded for the presence or absence of descriptions that minimally matched these categories. For example, we counted students’ responses, “Buy airtime,” “Pay networks,” or “Give big $$ to networks to air commercial” as acceptable responses to post-production step five, but we did not award points to a student who wrote, “Sell the ad to a network,” since this comment did not convince us that the student had an accurate understanding of the financial arrangements between advertisers and networks. To measure students’ knowledge of advertising production, students received a summed score representing the total number of production steps identified.

Measuring Media Literacy Skills

Following from the work of Quin and McMahon (1994) and Hobbs and Frost (2003), the media literacy measures used in this study assess students’ ability to identify:

  1. 1. the assumed target audience;

  2. 2. the specific techniques employed in the construction of the ad, including visual and verbal elements, symbolism, and graphic design;

  3. 3. the assumed message subtext or unstated message of the ad; and

  4. 4. the ad’s assumed purpose.

Although the readers of a media message rarely have knowledge of the creator’s actual target audience, subtext, or purpose of a specific media message, critical readers and viewers may make inferences about these dimensions of a media message using clues in the message text and images, along with their knowledge of media industries (Messaris, 1994).

To measure students’ assessment of presumed target audience, students were given a checklist of six different age range categories, (from age two to over age 60); two genders; five different racial categories; five different social class categories (from “poor people” to “wealthy”). They were asked, “Who is the target audience for this message?” and invited to check all that apply. Faculty who responded to these test items agreed that the categories of race, age, gender and social class were the most relevant variables in relation to this ad.

Students were expected to provide a rationale for their choices of target audience. An open-ended question asked students to provide a description of specific visual information from the ad that they used to support their answers. Responses were coded on a four-point scale. For example, students who wrote: “it shows two good looking young black guys wearing suits” received four points, while the response, “looking expensive” received only one point and “beer is for the working class” received no points. In this scale, students who were able to describe specific verbal or visual features of the media message to justify their assessment of the target audience demonstrated stronger analysis skills than students who used only their existing knowledge or attitudes without examining the design features of the ad. This scoring strategy reflects current scholarship in literacy education, which emphasizes the ability of students to connect their interpretive responses to textual features of a message as a key dimension of critical thinking (Scholes, 2001).

To measure students’ recognition of the constructedness of media messages, students were asked to respond in writing to this open-ended question: “List the attention-getting techniques used in the ad.” After reading a sample of student responses, a protocol was developed as described earlier to code the quality of student responses. Coders were provided with a list of acceptable answers and coded for the presence of items. For example, students might describe the large size of the beer can, the drops of condensation indicating its coldness, the slogan, the lettering and font choices, the facial features of the two men depicted in the ad, or the use of visual contrast and lighting. Students who recognized that an ad contains many different visual and verbal elements that are designed to attract attention and communicate meaning through symbol systems were considered to have stronger analysis skills than those who did not identify these construction techniques.

Students were asked to describe the subtext of the message, defined as the unwritten meaning that the print ad is trying to convey to the viewer. After reading a sample of student responses, a protocol was developed as described earlier to code the three most frequently identified subtexts that students identified: coolness, lifestyle change, and drinking responsibly. Dummy variables were created to code for the presence of these variables in students’ written responses. In addition, students were asked to comment on the meaning of the small logo placed on the lower right hand corner of the ad, which used the shape of a highway traffic sign and the words, “Think When You Drink.” Responses were coded on a four-point scale. For example, students who wrote, “Don’t drink and drive” received a lower score than students who wrote, “They want to sound responsible, plus it’s good PR.” The ability to recognize a legal, economic or public relations dimension in explaining the meaning of the logo was considered a demonstration of higher levels of critical analysis skills than only identifying the message’s connection to the problem of drunk driving. This scoring strategy reflects current arguments among advocates of media literacy who recognize the centrality of examining the political and economic contexts in which media messages are produced (Lewis & Jhally, 1999).

Students were asked to identify the purpose of the print ad in an open-ended question. Responses were coded on a scale to reflect the different levels of understanding of how persuasion in advertising works. Three main categories of responses were identified after reviewing a sample of student responses. Students who wrote answers like “to sell beer” or “to make money” identified the business purpose of advertising. Students who wrote answers like “to get people to want to drink beer” identified the persuasive purpose of advertising. Students who wrote “to link feelings of ‘coolness’ with beer drinking” identified the associative emotional purpose of advertising. This coding strategy is based on an argument developed by Messaris (1994) about the dimensions of image analysis in advertising. Students who described the ad’s purpose in terms of creating a specific emotional response in viewers that could be linked to the consumption of the product were considered have more sophisticated levels of understanding of the purpose of the ad than those who merely described the ad’s function to increase sales volume.

Research Results

Students in the treatment group who received the year-long program of media literacy instruction in Grade 11 were compared to a control group in a different school district who received only the pretest and posttest with no treatment. The data were analyzed using analysis of covariance (ANCOVA) with the Minitab statistical program. In this analysis, the pretest scores for each variable served as a covariate and the posttest scores were the dependent measures. Home computer use was included as a covariant in all analyses to reduce some of the variance that may be a result of population differences. The analysis of covariance provides an ability to control for initial differences in the two groups, which is a characteristic typical of quasi-experimental designs. It can also be used with unbalanced designs, when sample sizes between groups are unequal. Because pretest variables are usually highly correlated with posttest variables, the ANCOVA design reduces the variability in the posttest scores that are associated with the pretest scores. On all measures, tests for normality and homogeneity of the within-group regressions were conducted to satisfy the assumptions for the analysis of covariance.

Increased Knowledge of TV Advertising Production Processes

Students in the media literacy treatment group increased their overall knowledge of the production processes involved in constructing a TV commercial, with most of the gain generated by students’ increased knowledge of pre-production processes. For example, 14% of media literacy students included the step of identifying target audience(s) and marketing goal(s) as compared with only four percent of control group students, F (2, 737) = 4.78, p < .05. Similarly, a large majority of media literacy students (70%) recognized that planning and brainstorming were necessary to create an ad, while only 26% of control group students identified this step, F (2, 737) = 13.89, p < .001. Statistically significant differences were also found between media literacy students in their identification of the development of visual planning materials, including storyboards or other visuals (treatment M = 17%, control M = 10%), F (2, 737) = 8.64, p < .001. Among the media literacy students, 24% explained that the ad agency would seek client approval, as compared with only four percent of the control group students, F (2, 737) = 11.88, p < .001. Students participating in the media literacy treatment were also more likely to recognize some aspects of the economics of advertising, as 24% included a description of the process of payment for airtime on various channels and networks. Only five percent of control group students listed this step, F (2, 737) = 7.18, p < .001. Table 1 presents these data.

Table 1. Knowledge of Advertising Production Processes
Click for larger view
View full resolution
Table 1.

Knowledge of Advertising Production Processes

Increased Media Literacy Skills in Analyzing Advertising

Identification of Presumed Target Audience

Students in the media literacy treatment revealed a tendency to identify the target audience more narrowly as compared with the control group. In contrast, control group students were more likely to identify the ad as targeting the broadest range of consumers. A number of statistically significant differences between the treatment and control groups reveal this pattern. For example, in evaluating the age of the intended target audience, the media literacy group was less likely to identify the target audience as older. Only 24% of media literacy students identified the audience as composed of 40 to 60 year olds as compared with 33% of control group students, F (2, 737) = 4.82, p < .05. Similarly, 10% of media literacy students identified the target audience as over age 60, as compared with 42% of control group students, F (2, 737) = 11.32, p < .001. Students who received media literacy instruction were less likely to interpret the ad as targeting women consumers, F (2, 737) = 6.02, p < .01. Table 2 presents these data.

Table 2. Identification of Target Audience
Click for larger view
View full resolution
Table 2.

Identification of Target Audience

For the largely white, middle-class high school students who participated in this study, media literacy instruction seemed to have a meaningful impact on their interpretation of how racial identity and racial identification are used by advertisers as a component of message targeting. After receiving media literacy instruction, students were much less likely to identify the ad as targeted to whites than those in the control group. In other words, students in the control group were much more likely to claim that the ad was targeted to all racial groups. Students in the media literacy treatment group were more likely to identify the target audience more narrowly. For example, students in the media literacy group were less likely to see the message as targeted to whites (treatment M = 49%, control M = 60%), F (2, 737) = 11.52, p < .001. In addition, students who received media literacy instruction were far less likely to see the message as targeted to Hispanics ( F (2, 737) = 13.95, p < .001), Asians ( F (2, 737) = 8.19, p < .001) or other racial groups ( F (2, 737) = 14.27, p < .001).

In analyzing students’ assessments of the social class of the intended target audience, students who received media literacy instruction were also less likely to see the print beer ad as targeted to the poor. Only 24% of the media literacy students identified the beer ad as targeting poor people, as compared with 43% of the control group students, F (2, 737) = 10.43, p < .001.

Students who participated in media literacy instruction were more likely to support their reasoning about the assumed target audience by referring to specific design features of the ad. After identifying the intended target audience for the print beer ad, students were invited to provide open-ended justification of their reasoning for selecting the target audiences that they identified. As described earlier, these responses were coded for the number of examples of visual or verbal evidence from the text. Students who received media literacy instruction were much more likely to use more explicit description of visual and verbal elements of the ad in explaining their reasoning, (treatment M = 2.24, control M = 1.63), F (2, 737) = 10.59, p < .001. Table 3 presents these data.

Table 3. Advertising Analysis Skills
Click for larger view
View full resolution
Table 3.

Advertising Analysis Skills

Identification of Message Construction Techniques

Students in the media literacy treatment identified a greater number of message construction techniques than those in the control group (treatment M = 2.50, control M = 1.37), F (2, 737) = 7.28, p < .01. Differences in students’ open-ended responses were statistically significant, demonstrating students’ ability to identify a greater number of verbal, visual and symbolic elements of the ad that were designed to attract and hold viewer attention.

Identification of Message Subtext

ANCOVAs revealed statistically significant differences between the treatment group and the control group on all of the three subtext variables. Students in the media literacy treatment were more likely to identify the “be cool” theme ( F (2, 737) = 10.22, p < .001), with 51% of students identifying this subtext in the treatment group as compared with only 28% in the control group. Students in the media literacy treatment were more likely to identify the “change your lifestyle” theme ( F (2, 737) = 8.57, p < .01), with 47% of treatment group students identifying this theme as a compared to only 22% of the control group. In addition, statistically significant differences were also found between treatment and control groups in the identification of the “drinking responsibly” theme ( F (2, 737) = 9.34, p < .01), with 47% of media literacy students identifying this theme as compared with only 18% of control group students.

In analyzing the students’ responses to an open-ended question asking them to comment on the meaning of the small logo that read, “Think When You Drink,” there were statistically significant differences between control and treatment groups in the identification of this logo as part of a PR strategy. Thirty-one percent of students in the media literacy group identified a public relations motive as compared with 26% of control group students, F (2, 737) = 7.98, p < .01. There were no statistically significant differences between groups in the identification of an anti-drunk driving theme.

Identification of Purpose

Students in the media literacy treatment group increased their ability to identify the ways in which advertising persuades by linking emotions to the consumption of products. The proportion of treatment group students identifying the associative emotional purpose of the ad increased from a pretest mean of 15% to a posttest mean of 22%, F (2, 737) = 4.52, p < .05. However, because students in the media literacy treatment group had statistically significant differences at the time of pretest in their identification of the purposes of the print ad, tests of statistical significance between control and treatment groups could not be used. At the time of pretest, students in the control group were more likely to identify a business reason (84%) as compared with students in the media literacy treatment group (35%), and 45% of the media literacy group identified a persuasive purpose compared with only ten percent of the control group. However, these pretest differences were the only statistically significant differences found between groups at the time of the pretest, which further indicates the overall parity between groups.

Relationship between Media Literacy Skills and Advertising Knowledge

In order to examine the relationship between students’ knowledge of TV advertising production processes and their media literacy skills, two summed scores were created. As described earlier, an advertising production knowledge score was created by summing the total number of steps identified by students in describing the TV advertising production process. This score ranged from 0 to 17, with a control group mean posttest score of 7.83 as compared with treatment group mean score of 11.47. A media literacy advertising analysis score was derived by summing the scores for the sub-tasks described above, with a with a control group mean posttest score of 8.10 as compared with treatment group mean score of 11.28. Pearson product moment correlation scores of .202 reveal a statistically significant association between the two variables. Regression analysis was used to examine the relationship between variables, showing statistically significant differences between control and treatment groups on advertising analysis scores, F (2, 737) = 7.25, p < .001, but advertising production knowledge did not predict analysis scores, F (2, 737) = 3.44, p < .06.

Strengthening Critical Thinking Skills by Analyzing Advertising

This research examined this central question in the context of field research in two large public high schools: how does media literacy instruction, integrated within a year-long course in high school English language arts, affect the development of high school students’ advertising analysis skills and their knowledge of advertising production processes?

Compared to students in the control group, students who received media literacy instruction gained substantial knowledge of the production processes used in advertising. They particularly gained knowledge of the preproduction processes involved in identifying a target audience, brainstorming and planning, and using visual tools to design TV ads. They also demonstrated more knowledge of the economics of advertising.

The relatively substantial growth in students’ knowledge from pretest to posttest suggests that perhaps the English teachers who administered the media literacy curriculum may have emphasized knowledge of media production processes or had extensive knowledge themselves of how TV ads are constructed. It is unknown whether or not such knowledge is typical among secondary English teachers nationwide. Advertising practitioners and media scholars might cast a critical eye on the range and diversity of curriculum materials now used in public secondary schools to teach about advertising so as to identify future opportunities for creating teaching resources that might enhance educators’ ability to teach about the advertising industry in the context of media literacy education.

Students who participated in the media literacy condition displayed growth in their ability to critically analyze advertising. First, students gained a more sophisticated understanding of the concept of target audience, recognizing that a particular print ad is not intended to reach all possible readers or viewers, but is designed to communicate to a specific demographic group. In particular, students in the media literacy treatment appear to have gained an increased sensitivity to racial difference between minority groups. Many media literacy advocates believe that a more nuanced conceptualization of audience is a key dimension in supporting students’ growth as active, socially-positioned users of media texts (Goodman, 2003). As Branston (1991, 119) writes, “The concept of audience has the potential to develop understandings of the cultural power of media institutions in ways that won’t reduce down to ‘brainwashing’ or ‘ideology’ the variety of relations we enter into with texts.” Evidence from this research supports the argument, put forth by Giroux and Simon (1989) and Cortes (2000) that media literacy instruction may better help learners situate themselves and others in sociopolitical context.

Among the most educationally relevant findings, students in the media literacy group demonstrated the ability to support their judgments and interpretations concerning target audience using specific visual, verbal, and symbolic evidence from the advertising text. The ability to support arguments with evidence is one of the most central skills of literacy education at the secondary level. As a result, this study provides preliminary evidence that advertising analysis activities, when integrated into English language arts education, support the development of students’ critical thinking skills.

Increased ability to identify construction techniques provides evidence that media literacy instruction leads to higher levels of awareness of the constructed nature of a print ad. Buckingham (1998) points out how knowledge of media languages—the sets of codes and conventions that are shared between producers and audiences—cannot be seen as merely neutral. Tyner (1998) considers the ability to recognize the constructedness of various media texts to be the central concept in media literacy education. Such awareness, notes Tyner, is generated as students serve “cognitive apprenticeships” with teachers who emphasize the process of using active investigation to unearth meaning-making processes using the texts of everyday life, making inferences and predictions in ways that “make visible to novice learners those powerful problem-solving strategies and heuristics that more expert readers practice flexibly and strategically” (Lee, cited in Tyner, 1998, p. 177). Using texts (like advertisements) in an English classroom, where students can be expected to have greater social or linguistic prior knowledge, may help students to master analytic skills that contribute to internalizing expert reading practices.

Interestingly, while this study found an association between knowledge of TV advertising and advertising analysis skills, we did not find that knowledge of advertising production processes predicts increases in advertising analysis skills. This finding runs counter to Potter’s (1998) observation that media industry knowledge structures enhance media literacy skills. However, this finding may reflect a limitation of the methodology used to measure knowledge of TV advertising production processes or it may suggest that knowledge of advertising is independent of critical analysis skills. Given the preliminary way in which this relationship was explored, however, this evidence is tentative and suggests that further research should investigate the relationship between knowledge of advertising and critical analysis skills using more nuanced measures of advertising knowledge and more sophisticated theoretical models of the potential relationship between knowledge and critical analysis skills.

This study contributes to our understanding of how educators’ use of advertising in classrooms can activate adolescent critical thinking that promotes literacy development. As described by Scholes (2001, p. 215), “a proper craft of reading—including what we learn from reading poems and other literary works—can and should be used as an instrument for the serious study of all kinds of textual objects” (emphasis added). In this view of literacy, the craft of reading should involve the application of critical thinking skills in relation to a wide range of print and non-print texts, including advertising. Ads provide an important set of texts that can help students strengthen traditional literacy analysis skills, including identifying message design and construction techniques, recognizing how authors express specific values and points of view, comparing and contrasting messages with similar content, and noticing when information is omitted from a message. This study contributes to the field by showing how these skills are developed through everyday classroom instruction and how they are activated in the context of exposure to different media genres, particularly advertising. This study broadens the concept of cognitive defenses by examining more broadly some educational components of what might be termed media analysis skills, which use some key elements of the work of Robertson and Rossiter (1974) but are more closely aligned with approaches developed by British, Canadian and Australian educators who emphasized specific instructional practices designed to enhance students’ critical thinking about media messages (Film Working Group, 1999; Tyner, 1998; Quin & McMahon, 1995).

This study examined students’ interpretations of the purposes of a specific advertising message. As Scholes (2001, 230) points out, while scholars have argued about the impossibility of discovering the intention of an author, at the same time “we must seek an authorial intention, while recognizing there are many reasons why we shall never close the gap that separates us from the author. The crafty reader must seek authorial intention knowing that what is found will never be exactly that.”

This research evaluated a limited number of meaningful critical thinking skills in response to advertising by examining students’ skills as activated by a current educational initiative on the part of one school district to integrate the study of advertising into the secondary English language arts curriculum. The treatment was a program of studies designed by ordinary classroom teachers with minimal involvement on the part of scholars or experts. The instruction students received was not a specially designed, intensive, short-term experimental program implemented by university faculty or graduate students, but a course of study designed and implemented by regular classroom teachers whose specialty is not advertising, media production, or media literacy, but secondary English language arts. In addition, while the teachers shared some common instructional objectives, some common texts and activities, they were largely on their own to develop and implement the day-to-day work of the classroom. As a result, this research measures the impact of media literacy instruction on student skills as this instruction occurs in the real world, with all the variability that exists from seven teachers teaching over 300 Grade 11 students over the course of a full school year.

Another important contribution of this research is its innovative approach to measuring critical analysis skills in response to an advertising message. Paper-and-pencil measures were designed that allowed students to demonstrate analysis skills that Robertson and Rossiter (1974) identified as central to the development of cognitive defenses to advertising, operationalizing critical analysis skills to include the identification of purpose, target audience, subtext interpretation, and construction techniques.

As the first large-scale empirical work measuring the acquisition of advertising knowledge and message analysis skills among adolescents in a school-based field setting, this research provides some evidence that incorporating media analysis activities into the English language arts curriculum at the high school level can enhance students’ development in the critical thinking skills needed to strengthen cognitive defenses against advertising. Further research will be necessary to identify the mechanism(s) by which learning about media production actually shape the development of media analysis skills. And it will be important to identify those instructional practices that lead to the greatest increase in skills over time, since it is likely that various instructional practices can differentially affect the development of specific critical thinking skills. Most importantly, it will be necessary to measure whether the critical thinking skills used to analyze advertising while in school transfer to students’ media-consumption experiences outside the classroom walls.

Renee Hobbs

Renee Hobbs is one of the nation’s leading authorities on media literacy education. She is an associate professor of communication at Temple University’s School of Communications and Theater, where she directs the Media Education Lab at Temple University and co-directs the Ph.D. program in Mass Media and Communication. She also helped found the Alliance for a Media Literate America (AMLA), the national non-profit membership organization that hosts the National Media Education Conference. She won a Golden Cable ACE award for Know TV, a middle-school media literacy curriculum sponsored by the Discovery Channel. Her research examines the impact of media literacy on the development of students’ language development and critical thinking skills. She co-authored the best-selling secondary language arts textbook, Elements of Language (Holt, Rinehart, Winston, 2000) and has published in such journals as Reading Research Quarterly, American Behavioral Scientist, and the Journal of Communication.

She holds a B.A. in English Literature and Film/Video Studies, an M.A. in Communication from the University of Michigan, and an Human Development from Harvard University.


1. Editorial Note: This is a reproduction of the ad that was available to students, and not the color original.


Alvarado, M., & Boyd-Barrett, O. (1992). Media Education: An Introduction. London: British Film Institute.
Anderson, J. A. (1983). The theoretical lineage of critical viewing curricula. Journal of Communication, 30 (3), 64–70.
Austin, E. & Johnson, K. (1997). Effects of general and alcohol-specific media literacy training on children’s decision making about alcohol. Journal of Health Communication, 2 (1), 17–42.
Babbie, E. (1998). The Practice of Social Research. Belmont, CA: Wadsworth.
Baron, L. (1985). Television literacy curriculum in action. Journal of Educational Television, 11 (1), 49–55.
Boush, D., Friestad, M. & Rose, G. (1994). Adolescent skepticism toward TV advertising and knowledge of advertiser tactics. Journal of Consumer Research, 21, 165–175.
Branston, G. (1991). Representation. In David Lusted (Ed.), The Media Studies Book: A Guide for Teachers. London: Routledge.
Brown, J. A. (1991). Television “Critical Viewing Skills” Education. Hillsdale, NJ: Lawrence Erlbaum Associates.
Brucks, M., Armstrong, G. & Goldberg, M. (1988). Children’s use of cognitive defenses against television advertising: A cognitive response approach. Journal of Consumer Research, 14, 471–482.
Buckingham, D. (2003). Media Education: Literacy, Learning and Contemporary Culture. Cambridge, England: Polity Press.
Buckingham, D. (1998). Media education in the U.K.: Moving beyond protectionism. Journal of Communication, 48 (1), 33–43.
Christenson, P. (1982). Children’s perceptions of TV commercials and products: The effect of PSAs. Communication Research, 9 (4), 491–524.
Cook, T. & Campbell, D. (1979). Quasi-experimentation: Design and Analysis for Field Settings. Boston, MA: Houghton Mifflin.
Cortes, C. (2000). The Children Are Watching: How the Media Teach About Diversity. New York: Teachers College Press.
Dorr, A., Graves, S. & Phelps, E. (1980). Television literacy for young children. Journal of Communication, 30 (3), 71–83.
Gersten, R., Baker, S., & Lloyd, J. (2000). Designing high-quality research in special education: Group experimental design. Journal of Special Education, 34 (1), 2–18.
Giroux, H., & Simon, R. (1989). Popular Culture, Schooling and Everyday Life. Granby, MA: Bergin and Garvey.
Goldberg, M. & Bechtel, L. Adsmarts: An intervention invoking reactance to combat alcohol usage by youths. Pennsylvania State University. Unpublished paper.
Goodman, Steve (2003). Teaching Youth Media. New York: Teachers College Press.
Kennedy, M. (1997). The connection between research and practice. Educational Researcher, 26 (7), 4–12.
Kress, G. (2003). Literacy in a New Media Age. London: Routledge.
Kubey, R. & Baker, F. (1999). Has media literacy found a curricular foothold? Education Week, October 27. Retrieved November 20, 2001, from
Lewis, J. & Jhally, S. (1998). The struggle over media literacy. Journal of Communication, 48 (2), 109–120.
Livaudais, M. (2003). Personal communication with the Director of Language Arts, Division of Curriculum and Instruction, Texas Education Agency, Austin. January 7.
Messaris, P. (1994). Visual Literacy: Image, Mind and Reality. Boulder, CO: Westview Press.
Odell, L., Vacca, R., Hobbs, R., Irvin, J., & Warriner, J. (2000). Elements of Language. Austin, TX: Holt Rinehart, Winston.
Potter, W. (1998). Media Literacy. Thousand Oaks: Sage Publications.
Quin, R., & McMahon, B. (1995). Evaluating standards in media education. Canadian Journal of Educational Communication, 22 (1), 15–25.
Roberts, D., Christenson, P., Gibson, W., Mooser, L. & Goldberg, M. (1980). Developing discriminating consumers. Journal of Communication, 30, 229–231.
Ross, R, Campbell, T., Huston-Stein, A. & Wright, J. (1981). Nutritional misinformation of children: A developmental and experimental analysis of the effects of televised food commercials. Journal of Applied Developmental Psychology, 1, 329–347.
Rossiter, J. & Robertson, T. (1974). Children’s TV commercials: Testing the defenses. Journal of Communication, 24, 137–144.
Richardson, V. (1994). Conducting research on practice. Educational Researcher, 23 (5), 5–10.
Rogow, F. (2001). AMLA: A new national organization for media literacy. Community Media Review, Spring, 24.
Scholes, R. (2001). The Crafty Reader. New Haven: Yale University Press.
Scruggs, T. & Mastropieri, M. (1994). Issues in conducting intervention research: Secondary students. In S. Vaughn & C. Bos (Eds.), Research Issues in Learning Disabilities. New York: Springer Verlag, 130–145.
Slater, M., Rouner, D., Murphy, K., Beavais, F., Van Leuven, J. & Domenech-Rodriguez, M. (1996). Adolescent counterarguing of TV beer advertisements: Evidence for effectiveness of alcohol education and critical viewing discussions. Journal of Drug Education, 26 (2), 143–158.
Singer, D., Zuckerman, D. & Singer, J. (1980). Helping elementary school children learn about TV. Journal of Communication, 30 (3), 84–93.
Texas Education Agency (1998). Texas Essential Knowledge and Skills. Retrieved March 20, 2001 from
Tyner, K. (1998). Literacy in a Digital World. Mahwah, NJ: Lawrence Erlbaum Associates.
Vooijs, M. & Van der Voort, T. (1993). Teaching children to evaluate television violence critically: The impact of a Dutch schools television project. Journal of Educational Television, 19 (3), 139–152.
Young, B. (1990). Television Advertising and Children. Clarendon Press: Oxford, England.

Additional Information

Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.