Smoking Advertisements And Impact On Human Behavior Finance Essay - Free Essay Example

Consumers future decision and firms advertisements to induce it, has long history. An investigation of the five possible decision choices, an individual face in the society where there are both pro- and anti- smoking advertisements are competing to provide the information about the costs and benefits of the change in the behavior of the individual toward their favor by working on the cycle of exposure, recognition, receptivity and agreement phases of advertisement. Health Belief Model indicates, individual change their decision by working on this four step cycle. Hence there meant to be some association between this cycle and the future health related decision. Introduction Smoking is currently an issue that youth face nowadays; most of non-smokers are inspired from their surroundings and decide to start to smoke. This decision to smoke has included a lot of changing in the perceived impression about the cost of this behavior. Even though this perceived cost of smoking behavior is not directly measurable, it is only observable but still it is the part of information cloud that surrounds him. And the main suppliers of this information are advertisements. Hence in this era of advertisement the respondents have to decide about his options according to the information available, according to the context of consumer behavior, the amount of sales the product gets depend upon the exposure, recognition, receptivity and agreement the advertisement makes. (1) As these smoking related advertisements are of two types the advertisements that induce people to smoke (pro-smoking advertisements) and the advertisements that restricts the people to smoke (anti-smoking advertisements) both are forcing the individual to consider its health related behavior decision. In this case of health related decision making, model has been constructed to observe peoples decision about the health related behaviors, as a result of smoking advertisements. In this respect, the perceived threat and net benefits, perceived susceptibility, perceived severity, perceived benefits and perceived barriers are analyzed. See for details, Schiffman and Kanuk (2007) This paper attempts to explain the decision making, of an individual which faces, when as a result of, pro and anti smoking advertisements through the mixture of both patterns explained above. It means that a person will commit to smoke if the exposure, recognition, receptivity and agreement created by the pro smoking advertisement are greater than the anti smoking advertisements. The models only are seen as the exposure and the frequency of delivered threats or benefits. The recognition can be seen as wh en the individual knows the content that is being delivered, the receptivity is the attachment of the individual to the product of service and the agreement is the views of the individual about the perceived benefits or perceived threats about the product or service. Straub et al. (2003) evaluated the work of pro and anti smoking advertisement effect using four point Likert scales about the decision of intention to smoke. For the evaluation of decision behavior can be done using multinomial logit model as used by Nguyen (2003). Schiffman (2007) stated that effect of exposure and recognition is diferent on different individuals. There can be individuals with high exposure with low associated agreement or vice-versa. It might have different effect on future intnention for different type of people. Hence not the frequency but effective frequency(2) is expected to have any or more association with the future intentions. Similarly, there are some sources that contain strong sign ificant agreemrnt, as compared to the other sources are expected to have some correlation with the future intentions; it can be called as source strenght(3). As this both parameters are giving weightage to the agreement that it is generating, hence they are expected to be independent from the individual difference. A survey is designed to generate the frequencses of messages seen and sources used for both pro and anti smoking advertisments and this is checked this assiciation with the future intentions to smoke. The precedings of the paper contains the review of the work of other scholars relaed to this issue, discription about the sampling procedure, variables their measuring mechanism and the methodology. The estimatied results that are obtained using quantitive measures, its reprsentation and intrepretation, limitations related to the research, colclusion of the paper with policy recommendation and the references of the realated and used articles are presented hereafter. Effective frequency is generated by the sum of number of times the massage is delivered and the quantified scale of agreement with it. Source Sternght is generated by alloting it the quantified agreement value of the message that it is delivering Economics of Smoking: Higher the cost of a behavior, the less people will do it, while the lower the cost, the more people will do if it. (4) The decision to smoke depends upon the amount of information about the cost and benefit of this habit. Hence the positive smoking advertisers are providing information that is reducing the perceived cost of the smoking where as the anti-smoking advertisements are providing information that is increasing the cost of this behavior. People who smoke or ought to smoke think to reap satisfaction from the consumption of cigarettes. An anti-smoking activist Scott Ballin asserts that There is no positive aspect to [smoking]. The product has no potential benefits. (5) Hence those who pay for to smoke are willing and paying for it due to the unreal reduced cost of the smoking habit. Whereas those who have not smoked due to anti-smoking advertisements, had perceived higher cost of this habit i.e. smoking. This cost can be termed as the loss or damage to health that the smoking behavior causes. Williams, Walter E. Economics and Smoking. A Minority View. March 14, 2007. https://www.gmu.edu/departments/economics/wew/articles/07/smoking.html. Lernieux, Pierre. The Economics of Smoking. June 28, 2000. https://www.econlib.org/library/Features/feature5.html. Secondly this analysis of advertisements as an analysis of change in human health related behavior can be useful for the suppliers of information (advertisers) about the effectiveness and dimensions of the resources spent on these advertisements. Similarly as Government is responsible of increasing the welfare of the public, then it must have to put justifiable effort to restrict its public to not to indulge in health damaging activities. The effectiveness of the sources applied by the advertisers is important to be analyzed so that if one of the source becomes ineffective the advertisers can focus on the alternate resources to make the target public to make decision in their favor (i .e. decide to smoke for the positive advertisers). Literature review Generally, the non smokers and individuals face some choices as the trend to smoke is very common and there is a lot of advertisement both positive and negative going on in the society. The choices can be like completely deciding to smoke or not to smoke. According to Hochbaum et al. (1950s), humans decision to change his habit about smoking goes through four stages i.e. exposure, recognition, receptivity and agreement. In this research the dependent variable is the humans decision to the advertisements and the independent variables are as stated above. Both pro and anti advertisers follow somewhat specific pattern. The empirical evidence of the health belief model (6) was given by Doroodian (1991). The demand of cigarette was decreased by the increased anti smoking exposure efforts by the government and health associations in 1964 (Surgeon Generals Report), 1968 (effective years of fairness doctrine), 1971 (pro smoking broadcasting ban) and 1979 (2nd Surgeon Generals Report). The health belief model was spelled out in terms of four constructs representing the perceived threat and net benefits: perceived susceptibility, perceived severity, perceived benefits, and perceived barriers. These concepts were proposed as accounting for peoples readiness to act. The importance of the effect of exposure was again explored by McVey (2000) by treating the sample through four different methods; each having different level of exposure (using TV and local campaigns) and proved that greater the exposure more incidences of smoking. An experiment done through newspapers by Cummings et al. (1987) in the region of Buffalo, New York gave significant results in making people decide not to smoke. The other avenue from where the antismoking message can be delivered is the cigarette pack itself. Thrasher et al. (2007) compared the difference between the message printed with graphic imagery and without graphic imagery on the cigarette pack and came through with the conclu sion from the experiments in Canada and Mexico that graphic imagery decrease the demand of cigarette sold. The weakness of these all papers is that they havent checked the effect of these sources when they are moving with others sources, which is the case of real life. After the importance of exposure, follows the importance of recognition, it is the quality of message that is being delivered. The anti smoking message can be presented by two way one to present the cost of smoking and other is to present the benefit of non smoking. As Bajde (2008) suggests that the effect of anti smoking message is different to smokers and non smokers according to the type of message delivered. The significance of recognition was also evaluated by Siziya (2008) using GYTS (global youth tobacco survey) by identifying the positive association between current smoking status and the passages like direct advertisement, parents, peers, teachers and religious prohibition about smoking. Receptivity and agreement are very important in decision making. Pechmann (2006) worked on this phase by evaluating the results after delivering different type of messages to the audience. He concluded that advertisements showing the extreme consequences of the smoking were rewarded with the greater number of agreements. Increase in experience from campaigns, according to Popham (1993) can also affect receptivity of the individual to have a decision about the health behavior. In the end, aggregating all the stages Straub et al. (2003) used effect of both pro and anti smoking advertisements on the non smokers and evaluated their four liker scale intend to smoke decision and proved that the exposure to advertisements and agreement with the messages delivered plays its role in changing the future decision behavior. The limitation in this paper and all previous stated studies that they have not considered the fact that different dosage of advertisement and different sources have different effect on different individuals. Hence this paper is designed to find the association between the determinants of future intentions that are expected to be independent of difference in the individual behavior. It may be noted that there is hardly such study pertaining to Pakistan, particularly focused on young college graduates. It is the prime age in which individual gets addicted to smoking or equally important is that someone quits smoking. Thus, this study is a pioneering in terms of analyzing such an important behavior of youths which is concerned about their quality of life. Given the above rationale, this study aims at achieving the following objective: Identification of significant sources that deliver pro and anti smoking advertisements. Point out significantly different pro and anti tobacco messages which encourage or discourage smoking Find out future intention to smoke which is effected by the effective frequency and source strength of the pro and anti smoking a dvertisements Methodology Sampling Framework The sampling frame used for this study is the Students of Forman Christian College Lahore. A random sample technique is used. The size of that sample is calculated from the formula proposed by Parel et al. (1973) as stated below: n = Where; N = Population size from which the data is taken (3122), Z= normal variable (2.56 @ 99%) d = maximum error deemed acceptable (10%), p = proportion of the indicator in population calculated from survey (0.30) n = 114 Hence: From this formula using specified parameters the calculation the optimal sample size comes out to be 114. Hypothesis and Proportion Generation For the generation of the hypothesis of significant messages, a survey was conducted in which several pro and anti tobacco statements were asked. From the top six pro and anti tobacco messages were adopted for the study and the geometric mean of all their proportions is used as the proportion (calculation mechanism stated below) that will be used for sample generation formula. Here pi, is the proportion of the occurrence of the ith message and p is the mean of all the proportions which came out to be 0.30. After the calculation the average proportion that any message exists in the population is 0.30 which means that out of 100 people on average 30 people have heard about the message (positive or negative tobacco related). Questionnaire; A questionnaire was developed to generate data to be used for analysis. For convenience the questionnaire was prepared amended and revised in the light of comments by the experts (7). An optimal size of the questionnaire was adopted for the study. Variables / Indicators; The variables in the model are; Intentions to smoke Recognition with the source of the pro tobacco message Frequency of pro tobacco message delivered Agreement with the message; pro-smoking advertisements Receptivity of pro tobacco advertisements Recognition with the source of the anti tobacco message Frequency of anti tobacco message delivered Agreement with the message; Anti-smoking advertisements Receptivity of anti-smoking advertisements Awareness of laws and rules (social and organizational restrictions) (7) I am thankful to the comments on earlier draft to my advertisement expert of PERI, Lahore. Their comments helped to improve the study. Calculation mechanism of variables; Recognition with the source of the pro tobacco message In real life the advertisements had spillover effect that it increases awareness and deliver any particular type of message. Exposure (number of times which individual have seen the message) of anti-smoking advertisements measurement scale is previously used by GYTS (Global Youth Tobacco Survey) and adopted from Siziya (2008). In this calculation five different types of sources are discussed like TV movies, internet, shopping stores, newspapers magazines and billboards. Frequency of the pro tobacco message delivered A frequency of the message seen in a year is asked just to see the recognition like Straub et al. (2003) and Pierce et al. (1991). Higher the frequency represents the ease of recognition of that message. This information is used to form an indicator named Effective Frequency, which is expected to have positive relationship with future intention to smoke and greater association with it, as compared to t he original variable. This quantitative variable can be used in the form of qualitative variable. Receptivity of pro-smoking advertisements:- This measurement scale is adapted from Pierce et al (1991). This scale shows the response of the viewer is return of advertisements. There are two questions asked first one asking two scale responses about having bought any product with tobacco product logo on it and second one five scale question asking future chances of buying one product. The second question is expanded to five scale from four scale used by Pierce et al (1991) to give respondent a wider scope of answers for their future related action. The questions like what is your favorite brand? And what brand of cigarette will you prefer to buy? are dropped out of the model. The results are placed into the model in the form of individual variables. The sign of this variable is expected to have positive sign as present product buying can tempt the individual to buy tobacco prod uct in future. Both questions are dummy variables. Agreement with the message; pro-smoking advertisements The acceptance of the delivered messages is tested on the five scale statements. The agreement with every statement is entered into the model as separate variable. This concept is adopted as agreement is asked in the case of anti-smoking advertisements also, discussed later. Instead of measuring stress and depression like Straub et al. (2003), the statement like Smoking decreases stress and is a solution for depression is asked for agreement. Hence if the individual has a perception that stress and depression can be relieved by smoking then he will go for smoking whenever an individual feels stress or depression. The higher the agreement means value more the chances that the individual will opt to smoke; hence a positive sign are expected. It ranges from 1 to 5. Recognition with the source of the anti tobacco message Similar to the pro tobacco recognition variable this variable is also calculated against the anti tobacco messages. It is expected to have negative relation with the future intention to smoke. Frequency of the pro tobacco message delivered, The frequency of the anti tobacco message delivered is calculated and converted into the effective frequency of anti tobacco message and it is expected to have negative relation with the future intention to smoke and higher negative correlation as compared to the original variable. Receptivity of anti-smoking advertisements Just like pro-smoking case, the receptivity of anti-smoking is also tested with the question like have you ever attempted any kind of anti-smoking seminar, activity or campaign? for a two scale answer. The answers are placed in the model as it is. This variable will probably have negative sign. It is a dummy variable. Agreement with the message; Anti-smoking advertisements This variable adopted form Straub et al. (2003), which checks the extent of agreeme nt with the issues that are delivered by the anti-smoking advertisements. There are six statements having five scale options instead of four scale question as used by Straub et al. (2003) for the sake of wider scope. The results are placed in the model. It will be expected have negative sign and range from 1 to 5. Awareness of laws and rules (social and organizational restrictions) The awareness with the social and organizational restrictions can also make the individual to stop to commit to smoke. Thats why there are two questions like there is a fine when you are caught smoking in public places? And the institution where you belong is smoking free? thus two scale answer is demanded. This variable imposes a negative expected effect on the intention to smoke. It ranges from 0 to 1. Intentions to smoke, According to Straub et al. (2003) and his stated prior studies, intention to smoke is a valid predictor of smoking initiation. The adapted question from Straub et al. (200 3) is expanded to a five scale questions are asked. This variable is considered as the dependent variable. It ranges from 0 to 4 Data analysis mechanism ANOVA results(8) Following are the results of the separate ANOVA analysis on both Anti and Positive smoking advertisements against the sources of messages. Table 1. ANOVA of Positive Smoking Messages Source of variation Degree of freedom Sum of Square F F(5,25)= Sources 5 38617.3 39.43 3.13 Messages 5 550 0.56 3.13 Error 25 4896.6 Total 35 44064 Table 2. Mean variation in the sources Sources Mean Variation Difference Billboards 4.167 Newspapers and magazines 5 0.832 Shopping stores 8.67 3.67 Internet 25.83 17.16 Non media* 71.17 45.34 Movies and TV* 85.17 14 LSD = 19.8 at ÃÆ'Ã… ½Ãƒâ€šÃ‚ ± = 5% *Best two most significant sources (8) Calculation provided in appendix Form the table 1, where the degrees of freedom are the number of sources, messages and error respectively, it is seen that the positive smoking messages are not significant statistically and the sources of the positive smoking messages are statistically significant. Here to find the most different source of message the mean variation of each message is arranged in the ascending order in the table 2. For this, if the difference between the immediate two values of mean variation is greater than the value generated from the Least Significance Difference Test (9) then the particular source is reason that all the sources are become statistically significant. Hence it can be seen that the source named Movies and TV is statistically significant hence only this will be used in the regression analysis so that the limited sample size does not create any problem. Table 3. ANOVA of Anti Smoking Messages Source of variation Degree of freedom Sum of Square F F(5,25)= Sources 5 101671.47 17.29 3.13 Messages 5 3508.47 0.59 3.13 Error 25 29400.36 Total 35 134580.30 Table 4. Mean variation in the sources Source Mean variation Difference Billboards 21.67 Non media 35.17 13.5 Shopping stores 40.83 5.86 Internet 54.5 13.67 Newspapers and magazines* 120.5 66 Movies and TV* 169.17 48.67 LSD = 48.45 at ÃÆ'Ã… ½Ãƒâ€šÃ‚ ± = 5% *Best two most significant sources (9) LSD = t ÃÆ'Ã… ½Ãƒâ€šÃ‚ ±/2 (v) formula taken from Chaudhry (2009) From the ANOVA analysis of the anti smoking messages results are represented in the table 3. It can be seen that for this, the messages are again are not statistically significant and the sources are statistically significant and using the difference in the mean variation in table 4, Movies TV, and newspapers magazines are different from all, hence will be used in the regression analysis. Regression analysis A multinomial regression analysis will be done having change in future behavior, as a dependent variable. Its coefficients will represent the likeliness of change in the behavior. All the variables which passed through the ANOVA as significant are used in the regression analysis. For the judgment of behavioral type decision having limited and discrete outcomes a multinomial logit model as used by Straub et al. (2003) and Nguyen et al. (2003) will be applied. The model is given below; Y = f (significant sources of anti and pro-tobacco advertisements) Where, Y = future intention to smoke The dependent variable is the individuals future intention to smoke, and it is defined as following cases: Definitely not (y = 0) Probably not (y = 1) Dont know (y = 2) Probably yes (y = 3) Definitely yes (y = 4) Table no. 5 Parameter Estimates of Multinomial Logit Model Future decision to smoke a B Std. Error Wald df Sig. Exp(B) 95% Confidence Interval for Exp(B) Lower Bound Upper Bound Definitely not Intercept 1.623 .752 4.658 1 .031 Anti-smoking TV Movies .869* .326 7.117 1 .008 2.386 1.259 4.519 Pro-smoking TV Movies -1.42* .357 15.757 1 .000 .242 .120 .488 Probably not Intercept -.814 1.163 .490 1 .484 Anti-smoking TV Movies 1.89* .468 16.265 1 .000 6.613 2.641 16.562 Pro-smoking TV Movies -1.78* .434 16.718 1 .000 .169 .072 .397 Dont know Intercept -.231 .971 .057 1 .812 Anti-smoking TV Movies 1.10* .387 8.066 1 .005 3.003 1.406 6.416 Pro-Smoking TV Movies -1.01* .402 6.308 1 .012 .364 .166 .801 Probably yes Intercept -1.47 1.227 1.435 1 .231 Anti-smoking TV Movies .383 .380 1.013 1 .314 1.466 .696 3.089 Pro-smoking TV Movies .018 .483 .001 1 .970 1.018 .395 2.624 The reference category is: Definitely yes. *significant at 5% Number of observations = 114 Log-likelihood = 126.630 Chi-squared (8) = 53.728* Pseudo R2 = 0.154 Figure 1 Discussion of Results The interpretation of multinomial logit model is very difficult hence it is only interpreted by checking its marginal effects. The estimated regression results represent that the pro and anti smoking advertisements form TV and Movies are significantly describing the outcome definitely not, probably not and dont know (y = 1, y =2 and y = 3). For others it is insignificant which represents that they are alone not enough to induce individual to make decision to smoke in future. The other factors can be the non significant sources that are excluded from the model. The R2 is 0.152 showing that only 15% of the variation in the dependent variables is explained by the independent. For a Multinomial model the overall significance is represented by the Chi-squared value this is also significant at 1%. In this model the Definitely Smoke decision to smoke is taken as reference point because the sample public is non-smokers and their decision to smoke in future is being tested against the frequ ency of anti and pro smoking advertisements on TV and movies. The results represent that if the average frequency of the anti smoking advertisement on TV and movies is increased by one level (i.e. increasing from never to sometimes or sometimes to most of the time) the probability of individual to choose Definitely not to smoke, Probably not to smoke, Do not know and Probably yes to smoke will be increase by 2.49%, 2.70%, 0.86% and -1.1% on average (calculated by summing the slope value with the intercept) keeping other factors remain constant represented in the figure 1. Similarly if the average frequency of the pro smoking advertisement on TV and movies is increased by one level (i.e. increasing from never to sometimes or sometimes to most of the time) the probability of individual to choose Definitely not to smoke, Probably not to smoke, Do not know and Probably yes to smoke will be increase by 0.20%, -2.60%, -1.24% and -1.3% on average keeping other factors remain constant. The results of the pro smoking advertisements are different from the expectation is due to two reason, the first one is that the target sample is non smokers and according to this characteristic the pro smoking advertisement will generally have minimal effect on them. Secondly the pro smoking advertisement in the significant source of advertisement i.e. TV and movies is already very rare, most of the countries had banned it. Conclusion and Policy Implication After constructing a survey and questionnaire according to the health belief model (i.e. prediction of future health related behavior by evaluating degree of significance exposure, recognition, agreement and receptivity of the pro- and anti- tobacco advertisements) it is concluded that the health related behavior is only partially affected by delivering anti- or pro-tobacco messages from TV and Movies. Hence to stop non-smoking youth to indulge in smoking, these sources should be used primary tool. The role of the anti smoking advertisements is to increase the perceived cost of the smoking behavior in the youth. Future intention to smoke means that in present either there is a problem or the pro-tobacco advertisements are successful as compared to the anti- tobacco advertisements. The issue related to the paper is there is no way to test the truth of the correspondent and as the questionnaire is asking the frequency of the advertisements seen lately, it cannot be sure that inter pretation the frequency of the advertisements is same for all ,also it is hard to infer on unreliable memory recall based questions. The results of the questionnaires that are filled cannot be compared with each other, to overcome this particular problem the variables in same section are not used as a mean they are used individually. Form the results it can be seen that advertisements and messages are not an important factor that determines the future decision to smoke among all the sources of advertisements, the firms (advertisers) which use only TV Movies are marginally significant. Thus it can be said that the money spent by the firms on other sources is not much gainful in the scenario of graduate students of Forman Christian College University. However, it may be a special case, since F.C.C University has stick rules and awareness against smoking. It is a special case which may not be applied in general. In real life the future related decisions are always scenario relat ed, hence even an individual who has checked to not to smoke in future might smoke if the present scenario changes. There should be a shorter way to ask and detect the determinants of future smoking intention that do not involve the time variation and situation variation. The length of the questionnaire in any form of primary analysis is always unwelcomed by the respondents so the length of the questionnaire is minimized as much as possible.