Multimedia Science Education on Drugs of Abuse
By Twombly, Eric C Holtz, Kristen D; Tessman, Greta K
Dear Editor: Adolescent substance abuse is a significant public health problem in the United States. According to the National Survey on Drug Use and Health (NSDUH, 2006), 2.7 million youth used alcohol for the first time and 1.5 million used an illicit drug in the past year. Similarly, the 2005 Monitoring the Future survey found that 41 percent of eighth graders have used alcohol, and 21 percent have used an illicit drug (Johnston et al., 2006). These usage rates are alarming because drug use in adolescence can have serious physical, emotional, and social consequences. Physical consequences of adolescent drug use include short and long-term negative effects on the brain and body, including disease, impaired judgment, addiction, and even death. Risks for youth that drink or use drugs include high rates of depression and suicidal thoughts, a higher risk of later substance abuse and dependence, poor academic performance and school drop-out, early sexual initiation, and perpetration of violence.
Literature suggests that school-based substance abuse prevention programs can reduce drug use (McBride, 2003). In particular, increasing knowledge about the negative effects of drugs may effectively delay the onset of use and prevent negative consequences of use among youth. Evidence supports implementation of prevention curricula before high school, as by that time the majority of youth are already experimenting with alcohol and drugs (Foster et al., 2003). Results of the 2005 Youth Risk Behavior Survey suggest that 26 percent of the high school students in the U.S. started drinking alcohol before age 13, and each successive cohort of high school freshmen is initiating drinking at an earlier age (CDC, 2005). Early prevention is crucial, because delaying the onset of drug use during adolescence may delay or prevent future addiction and negative consequences (Grant et al., 2005).
Despite the benefits of school-based prevention, the amount of this instruction that children receive declined significantly in recent years. Indeed, with the implementation of the No Child Left Behind law, the emphasis on core academic subjects such as math, science, reading, and writing often means less instructional time for health and prevention. In this new educational climate, science education offers an opportunity to inject persuasive health information into the core curriculum. Because science education curricula are matched to state and national standards of learning, they can be integrated into the standard science curriculum, a core area less prone to cutbacks than prevention or health in a timepressed, standards-driven educational climate.
Because there is a negative correlation between the perceived risks of using a drug and actual drug use, a critical component of prevention is information about drugs and their consequences (Johnston et al., 2006). But science education differs from prevention in that this information is presented without admonitions against use, and drugs are not described as bad or dangerous. Despite the differences between science education and prevention, the presentation of information in a science education curriculum about drugs may be persuasive and may therefore change attitudes and intentions. In this manner, science education can be a type of "stealth" prevention in which students are presented with facts that are likely change their attitudes and behaviors regarding drug use, but not direction to do so.
The scientific concepts related to drug use, including information on brain function and neurotransmission and how drugs change these processes, are complex. Nonetheless, children of all ages may be able to master complex concepts when information is presented in an engaging and age-appropriate format. Multimedia applications such as CD-ROM-based games, where difficult concepts are illustrated and interactive and competitive elements are included to engage the learners, hold promise as education and prevention tools.
Interactive multimedia tools combine text, graphics, and sound in formats that can be manipulated by the user, and their use can actively engage students in the learning process and support children with different learning styles. Moreover, the interactive nature of multimedia products can encourage active learning and participation and holds particular promise for inquiry-based instruction, where students gather, analyze and synthesize information to solve multifaceted problems. What is more, interactive multimedia may provide students with opportunities to practice skills and may enhance their motivation and self-efficacy to make healthy decisions.
Based upon the scope and consequences of youths’ drug use, recent changes in the educational climate, and the potential power of multimedia tools to engage and educate, we designed and implemented an evaluation of a science education-based multimedia curriculum on drugs of abuse. The curriculum, entitled Keys to Brain Power!, is a federally-funded product that provides students in grades 6-8 with interactive lessons on the normal functions of the brain, nervous system, and the body, and how drugs change these processes. The curriculum’s goal is to increase knowledge about these topics and promote protective attitudes about drugs. Because of the public policy push hi K-12 education to increase the proficiency of students hi math, reading, and science, the curriculum content was aligned with National Science Education Standards (NSES) and key state standards of learning.
There are six lessons in the multimedia curriculum. Each focuses on a discrete topic and contains several interrelated activities, including quizzes, animated tutorials, and games. Lessons build on one another, where early lessons on the typical functioning of the brain serve as a foundation for later modules on how drugs change that functioning. The curriculum provides scientific information on many specific drugs, including alcohol, nicotine, inhalants, prescription and over-the-counter drugs, marijuana, cocaine, heroin, steroids, methamphetamine, and club drugs such as GHB, MDMA, Ketamine, and Rohypnol. All lessons were field tested with the target audiences prior to initiating the formal evaluation of the curriculum.
The multimedia curriculum was delivered through a self-contained CD-ROM. For the evaluation, the teacher provided a brief introductory discussion on the topic of the lesson and instructed students, working together in small groups, to explore the CDROM lesson at a self-directed pace. Students completed one lesson per 45- minute class period, over the course of two weeks, at a rate of three lessons per week.
The curriculum was predicated on the Theory of Reasoned Action (TRA), which posits that exposure to new persuasive information causes progressive changes in knowledge, attitudes, and ultimately behavior (Ajzen, 1991). People generally have negative attitudes toward behaviors that they believe will result in negative outcomes. This curriculum builds knowledge about the physical risks of drugs use, which theory suggests will then translate into attitude, intention, and eventually behavior change. A seminal work applied the TRA to drug abuse prevention (Fishbein & Middlestadt, 1987) and since that time has been used extensively to explain motivations for drug consumption (Morrison et al., 2002) and to design interventions to prevent drug use (Worden & Slater, 2004).
Although the TRA is common to the field of public health and prevention, it has not been applied to science education, which instead focuses on knowledge acquisition and application of the scientific process. The integration of behavior change theory and science education creates opportunities for schools to integrate prevention into the core curriculum as opportunities for traditional prevention become limited. Using multimedia to present science education material may further decrease barriers to adoption faced by traditional prevention curricula.
To test this theoretical framework, the evaluation used a quasi- experimental, pretest/post-test design. Eight charter schools located in North Carolina, Tennessee and Texas participated. Students in four of the schools were randomly assigned to an experimental group and received the multimedia curriculum. Students in the other schools were assigned to the control group, and these students received no exposure to the curriculum and only participated in the standard science curriculum in their school. Multiple classrooms in each school participated.
The evaluation was guided by several empirical questions. First, how did knowledge of science and drugs change from pretest to post- test? second, to what extent does group assignment (experimental/ control) relate to change in knowledge about science and drugs? In other words, what is the effect of the multimedia curriculum on knowledge change? Third, how do factors, such as protective attitudes toward drugs, relate to change in knowledge? Fourth, what is the statistical effect of the curriculum on knowledge change, when controlling for other key factors, such as grade, gender, and race of the students? To address these questions, we created several instruments to assess students’ knowledge and attitudes. For example, we developed a 24-item multiple-choice instrument to assess children’s knowledge about drugs and drug abuse before and after the curriculum intervention. Items directly test knowledge, application, and synthesis of content material in the curriculum. Moreover, we created a 5-item attitude survey that asked students to rate, on a 5point Likert scale, how likely negative social, academic, legal and health outcomes were from drug use. These measures were based on existing measures in the literature, and none exceeded a fifthgrade reading level, as tested by the Flesch-Kinkaid Readability Index. The pretest and post-test were identical, and the time between pretest and post-test in both groups was four weeks.
Additional independent variables include group assignment (experimental/control) and student-level demographic controls, such as gender, race, and grade. Experimental group assignment (yes/no) and gender (male/female) are measured dichotomously. Race is measured as white, African-American, and other. Grade is measured categorically.
We used both descriptive and multivariate statistical methods to analyze the data. But the manner in which the data were collected creates some methodological challenges and limitations to this study. Because we were not provided with unique identifiers for students in the study, we analyzed the data as separate populations by group assignment (experimental/control) and time (pretest and post-test). In addition, how schools were assigned to experimental and control groups created some participant bias. Indeed, there were significant differences in the population characteristics by age, grade, race, and location at pretest. These variations were also evident at post-test, suggesting that random assignment was not effective in creating equivalent groups. While we attempt to control for these differences, the generalizability of these findings to other school populations should be viewed cautiously. Still, the data consist of 1,342 observations from two points in time (pretest/ post-test), providing a well-sized data set to examine the extent of knowledge and attitudinal change in the two populations.
There are other limitations to this study. For example, because data were only collected to examine knowledge and attitude changes, the study cannot assess the effect of exposure to the curriculum on changes in behavior regarding the use of alcohol and drugs of abuse. As a result, this study focuses solely on the first step in the three-fold TRA process, namely, that the supply of information may shift knowledge about the relationship between the use of alcohol and illicit drugs and the physiological impact on the brain and body. Subsequent data collection will allow us to further test the tenets of the TRA, including behavior change. Moreover, data limitations prevent us from empirically testing other factors related to the TRA, such as the effects of social norms on knowledge acquisition and attitude and behavior change. Future data collection will also allow us to test for other direct or mediating factors.
The data reveal several key findings about the impact of the multimedia curriculum on knowledge and attitudes in the target population. First, the experimental group gained considerable knowledge between the pretest and the post-test. Indeed, differences in knowledge in the experimental group were positive and statistically significant in 14 of the 24 survey questions. Most impressive was the knowledge increase on the question of which part of the brain is likely to be damaged by teen alcohol use. At pretest, 19.9 percent of students in the experimental group correctly answered that the hippocampus had the greatest probability of being damaged. At post-test, 50.5 percent of survey participants identified the correct answer. Similarly, only 16.1 percent of participants knew at pretest which rat should be given a placebo in an experiment that tests drug effects. At post-test, the percentage of correct responses had significantly increased to 49.2 percent.
The composite score for the experimental group also demonstrates a significant difference in knowledge across the span of survey questions. As noted above, the composite knowledge score equals average number of questions answered correctly. On average, students correctly answered roughly 40 percent of the knowledge questions at pretest, but at post-test, the average correct response rate was 52.1 percent.
The experimental group did lose ground on two knowledge questions, although only one showed a statistically significant difference. From pretest to post-test, students exhibited a slight loss in knowledge (0.7 percentage points) about which drug is used by athletes to build muscle but can cause stunted growth (steroids). Still, even with the knowledge loss, four of every five participants in the experimental group correctly answered the question. More problematic is the considerable decrease in correct responses on the question of whether nonmedical use of prescription drugs over a long time can lead to addiction. From pretest to post-test, the percentage of experimental students correctly answering "sometimes" declined from 30.4 to 14.6. The explanation for this decrease is unclear and may reflect some ambiguity in the curriculum. Nevertheless, the students who answered the question correctly on the pretest but wrong on the post-test were more likely to switch their answers to "yes" than "no," potentially suggesting that they were developing more protective attitudes about drugs.
Participants in the control group saw very little change in knowledge from pretest to post-test. Students in this group exhibited statistically significant knowledge increases on only four of the 24 questions. Conversely, they exhibited significant knowledge loss on two of the 24 survey questions. And on average, students in the control group correctly answered 37.3 percent of the questions during the pretest and 37.1 on post-test, reflecting a small, expected change. This finding suggests not only did the students in the experimental group gain more knowledge across the two points in time, but these students also finished the study with significantly more knowledge about the effects of alcohol and drugs than did control group students.
In contrast to knowledge, the curriculum appears to be relatively ineffective in changing attitudes about drugs of abuse in the experimental group. Indeed, there were no significant differences – either positive or negative – on a series of attitudinal measures about alcohol and drugs. While not statistically significant, the biggest increase in protective attitudes related to students’ assessment of whether a person who uses drugs can do poorly in school or at work. At post-test, roughly 66 percent of participants in the experimental group answered "very likely" on this measure, compared to 59 percent at pretest. Still, the findings reveal that, although the attitudes failed to change in any significant way, students exhibited relatively strong protective attitudes about drugs of abuse at both points in time. Of a maximum of 100 percentage points that a student could receive for protective attitudes on the five survey measures, the average student earned a score of 86.5 at both pretest and post-test. In contrast, children in the control group showed a relatively steep decline in protective attitudes about alcohol and drugs on all five attitudinal measures. It is unclear why these drops occurred, but they may relate to testing fatigue.
The demographic and locational biases in the survey populations necessitate the need to control for these factors using OLS models in order to estimate the statistical impact of the curriculum on knowledge differences. The multivariate findings suggest that participation in the experimental group at pretest is significantly and positively related to knowledge scores, when controlling for other demographic and fixed effects. In fact, students in the experimental group provided nearly more one correct answer at pretest (Bela=0.78) on the survey than those in the control group, when holding constant other factors. Compared to their reference groups in the model, students aged 11 and 12 and those in grades 7 or 8 also scored significantly higher on the knowledge test, while students in North Carolina scored lower.
Compared to the model at pretest, the relationship between experimental group participation and knowledge was much higher at post-test. There, students who received the curriculum intervention produced nearly four more correct answers (Beta=3.86) than control group participants. And not only did the benefit of being a member of the experimental group gain in strength, but the effects of various demographic factors decreased. While students in grades 7 and 8 scored higher than those students in sixth grade, the significant relationship between age and knowledge, and location and knowledge, disappeared from pretest to post-test in the OLS models. In the end, the post-test model suggests that, holding certain factors equal, participation in the experimental group and receipt of the curriculum serves as the strongest determinant of knowledge about the effects of alcohol and drugs on the body.
Overall, the results suggest that the multimedia curriculum under evaluation is relatively effective at increasing knowledge about alcohol and drugs of abuse. In fact, there was a significant increase from pre- to post-test in overall knowledge about the topics in the curriculum in the experimental group. Furthermore, the significance of this knowledge change remained high even after effects of grade, geographic location, and ethnicity were controlled for using multivariate statistical techniques. In contrast, the control group showed no significant change in knowledge. The potential practical benefits of the curriculum are notable. Because school-based prevention is related to decreases in adolescent drug use, a multimedia curriculum that provides relevant information to youth may be an effective method to reach students. Indeed, the research suggests that a multimedia curriculum such as the science- based approach described in this paper is likely to engage students. In addition, the curriculum offers students an opportunity to build computer literacy while being exposed to health promotion and risk reduction content.
Moreover, aligning the curriculum with state and national science standards may increase the willingness of schools and their teachers to implement the curriculum. In fact, the curriculum is likely to be attractive to schools that are exploring ways to integrate health education into the core curriculum. Teachers may also find the curriculum attractive because its preparation demands are low, and their students can move through the lessons in an independent manner.
The evaluation did reveal a potentially important limitation of the curriculum, however: there was no statistically significant evidence that the curriculum positively effected attitudes about drugs of abuse. Although potentially problematic, the finding raises a number of possible explanations. First, the five-point measure on attitudes used for the evaluation was very limited and may have lacked the sensitivity to uncover the full range of attitude change. second, students in both the experimental and control groups entered the study with relatively strong protective attitudes toward drugs, suggesting a possible ceiling effect. Third, an immediate post-test may not have provided enough time for attitudes to change. Educational materials designed to change attitudes in youths must not only provide accurate information about the topic, but they should also encourage youths to apply the knowledge to their own lives. The multimedia curriculum provides the former, but the design of the evaluation may not have allowed time for students to apply the new information to their own lives. Attitudes are primarily formed and maintained through experiences in the environment, which may make them difficult to change in a very short time.
Taken together, the results of this evaluation provide preliminary support for the approach of using multimedia science education as a health education and prevention tool, at least with regards to changing knowledge about drugs of abuse. Whether through the use of a more sensitive attitudinal measure or by allowing more time to elapse before the post-test, more information is needed to determine the extent to which knowledge and attitude changes, as they relate to the curriculum, are connected. Finally, the results of the evaluation point to the need for research on other innovative approaches that decrease barriers to delivery of critical health information to students in the current educational climate.
REFERENCES
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179-211.
Centers for Disease Control and Prevention [CDC]. (2006). Youth risk behavior surveillance-United States, 2005. MMWR 2006, 55.
Fishbein M. & Middlestadt, S. E. (1987). Using the theory of reasoned action to develop educational interventions: applications to illicit drug use. Health Education Research, 2(4): 361-371.
Foster, S. E., Vaughan, R. D., Foster, W. H., & Califano, J. A. (2003). Alcohol consumption and expenditures for underage drinking and adult excessive drinking. Journal of the American Medical Association, 289(8): 989-995.
Grant, J. D., Scherre, J. F., Lyons, M. J., Tsuang, M., True, W. R., & Bucholz, K. K. (2005). Subjective reactions to cocaine and marijuana are associated with abuse and dependence. Addictive Behaviors, 30(8): 1574-1586.
Johnston, L. D., O’Malley, P. M., Bachman, J. G. & Schulenberg, J. E. (2006). Monitoring the Future 2006: Overview of Key Findings. University of Michigan News and Information Services: Ann Arbor, MI. [On-line]. Retrieved from: www.monitoringthefuture.org.
McBride, N. (2003). A systematic review of school drug education. Health Education Research, 18(6): 729-742.
Morrison, D. M., Mar, C. M., Wells, E. Z., Gillmore, M. R., Hoppe, M. J., Wilsdon, A., Murowchick, E. & Archibald, M. E. (2002). The theory of reasoned action as a model of children’s health behavior. Journal of Applied Social Psychology, 32(11): 2266-2295.
National Survey on Drug Use and Health [NSDUH]. (2006). Substance Abuse and Mental Health Administration. Retrieved from http://www. oas.samhsa.gov/prescription/toc.htm
Worden, J. & Slater, M. (2004) Theory and practice in the National Youth Anti-drug Media Campaign. Social Marketing Quarterly, 10(2): 12-27.
Eric C. Twombly, Ph.D., Andrew Young School of Policy Studies, Georgia State University, Atlanta, GA
Kristen D. Holtz, Ph.D., KDH Research & Communication, Atlanta, GA
Greta K. Tessman, Danya International Inc., Silver Spring, MD
Authors Note: This research was supported by funding from the National Institute on Drug Abuse, N44DA-3-1103.
Copyright American Alcohol and Drug Information Foundation Apr 2008
(c) 2008 Journal of Alcohol and Drug Education. Provided by ProQuest Information and Learning. All rights Reserved.