Health Promotion International, Vol. 30 No. 3 doi:10.1093/heapro/dat072 Advance Access published 22 October, 2013
# The Author (2013). Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
Reliability and validity of television food advertising questionnaire in Malaysia 1
Faculty of Health Sciences, Universiti Teknologi MARA, Puncak Alam Campus, 42300 Puncak Alam, Selangor, Malaysia and 2Nutrition Division, Ministry of Health Malaysia, Level 1, Block E3, Parcel E, Precint 1, Federal Government Administrative Center, 62590 Putrajaya, Malaysia *Corresponding author. E-mail: [email protected]
SUMMARY Interventions to counter the influence of television food advertising amongst children are important. Thus, reliable and valid instrument to assess its effect is needed. The objective of this study was to determine the reliability and validity of such a questionnaire. The questionnaire was administered twice on 32 primary schoolchildren aged 10– 11 years in Selangor, Malaysia. The interval between the first and second administration was 2 weeks. Test– retest method was used to examine the reliability of the questionnaire. Intra-rater reliability was determined by kappa coefficient and internal consistency by Cronbach’s alpha coefficient. Construct validity was evaluated using factor analysis. The test– retest correlation showed moderate-tohigh reliability for all scores (r ¼ 0.40*, p ¼ 0.02 to r ¼ 0.95**, p ¼ 0.00), with one exception, consumption of fast
foods (r ¼ 0.24, p ¼ 0.20). Kappa coefficient showed acceptable-to-strong intra-rater reliability (K ¼ 0.40– 0.92), except for two items under knowledge on television food advertising (K ¼ 0.26 and K ¼ 0.21) and one item under preference for healthier foods (K ¼ 0.33). Cronbach’s alpha coefficient indicated acceptable internal consistency for all scores (0.45–0.60). After deleting two items under Consumption of Commonly Advertised Food, the items showed moderate-to-high loading (0.52, 0.84, 0.42 and 0.42) with the Scree plot showing that there was only one factor. The Kaiser –Meyer –Olkin was 0.60, showing that the sample was adequate for factor analysis. The questionnaire on television food advertising is reliable and valid to assess the effect of media literacy education on television food advertising on schoolchildren.
Key words: reliability; validity; food advertising; schoolchildren
INTRODUCTION Childhood obesity is a global problem and it is a public health concern in Malaysia. The National Health and Morbidity Survey (NHMS) III reported that 5.4% of children aged ,18 years were overweight (weight-for-age . þ2SD) (MOH, 2008), whereas the reanalysed data of the NHMS III showed that 11.4% children aged 6–12 years were obese (BMI-for-age . þ2SD) (MOH, unpublished data). Whilst many factors contribute to obesity, food advertisements targeted at children has been recognized as one of the main factors, and
it is considered as an important arena of action in the prevention of obesity (WHO, 2006). Food industries are using various strategies to get to the children, and the most dominant medium used is television (Cairns et al., 2009). This is because television is the most effective way of marketing foods worldwide (Hawkes, 2004). Studies done by 13 independent research groups in 11 countries found that 18% of the television advertisements were for food, and overall, food was the second most frequently advertised product, after channel promotions (23%). Majority of food advertisements were for 523
Downloaded from http://heapro.oxfordjournals.org/ at Western Michigan University on March 25, 2016
ABDUL RAZAK ZALMA1,2*, MD. YUSOF SAFIAH1, DANIS AJAU1 and MD. ISA KHAIRIL ANUAR1
524
A. R. Zalma et al.
food advertising, which will be used to assess the effect of media literacy education on television food advertising among schoolchildren. Since food advertising influences children’s nutritional knowledge, food preferences and food consumptions (WHO, 2006), the questionnaire was developed to cover these three main dimensions which were knowledge on healthier food, preference for healthier food and consumption of commonly advertised foods on television. Knowledge on television food advertising was also included to gather some basic data on the children’s media literacy. METHODS This was a pilot study of a questionnaire intended to be used in a study to evaluate the effect of media literacy education on television food advertising on schoolchildren. The study protocol was approved by the Ethical Committee of Universiti Teknologi MARA Malaysia. Sample For this study a primary school in Selangor (one of the states in Malaysia) was selected and the participants were all the pupils from one class that had been selected at random. At the first administration of the questionnaire, 32 participants aged 10–11 years took part, and at the second administration 31 took part. Television food advertising questionnaire The questionnaire was developed to cover the three areas that can be influenced by food promotion to children (WHO, 2006). These are children’s nutritional knowledge, food preferences and food consumption. The other areas, which were purchasing and purchase-related behaviour and diet and health status, were not covered in this questionnaire. Items for knowledge on television food advertising (section E) were adapted from Jolls (Jolls, 2011), whereas items for knowledge on healthier food (section F) and preference for healthier food (section G) were adapted from Gwozdz and Reisch (Gwozdz and Reisch, 2011). Items for Consumption of Commonly Advertised Foods (section H) were adapted from Stansfeld et al. (Stansfeld et al., 2003). Sections A –D gather basic information of the children and their parents which are the parents’
Downloaded from http://heapro.oxfordjournals.org/ at Western Michigan University on March 25, 2016
non-core (unhealthy) food (67%), such as fast-food restaurant meals (12% of food advertisements), chocolate and confectionery (12%) and high-fat, high-sugar or high-salt spreads and sauces (8%) (Kelly et al., 2010). In a study conducted in Malaysia, it was found that 54162 advertisements were broadcasted by free-to-air TV stations during children’s prime time, and 27.9% composed of food products advertisements (Karupaiah et al., 2008). Snack foods which are considered as unhealthy food was found to be the most advertised food groups (34.5%), followed by dairy products (20.3%), confectionary (13.4%), biscuits (11.2%), fast food (6.7%), breakfast cereals (6.4%) and beverages (4.1%). Children need to be protected from heavy marketing of unhealthy food through television food advertising because marketing generates positive beliefs about advertised foods and influences children’s food preferences, purchase requests and consumption (Livingston and Helpers, 2004; Cairns et al., 2009; Nazari et al., 2011). The influence of food marketing is beyond their understanding of its persuasive intent (WHO, 2012). Harris and Bargh (Harris and Bargh, 2009) had proposed media literacy education as one of the three broad categories of solutions to protect children against the unhealthy influence of television and food advertising. Media literacy means the ability to gain access, analyse, evaluate, create and participate with media in all its forms (Jolls, 2011), thus media literacy education means teaching the skills associated with media literacy. Critical viewing skills acquired through media literacy education has been found to be related to higher taste ratings for healthy foods and lower taste ratings for unhealthy, highly advertised foods (Harris and Bargh, 2009). Thus, teaching children to defend against advertising influence through media literacy education is recommended to increase their critical viewing skills and scepticism about the media and advertising. Media literacy education on television food advertising is relatively new in Malaysia and to our knowledge there is no standard instrument developed for our population to determine the effectiveness of media literacy education. Other studies (Stansfeld et al., 2003; Gwozdz and Reisch, 2011; Jolls, 2011) regarding food advertising and food consumption have used instruments suitable for European and American population. The objective of this study was to examine the reliability and validity of a questionnaire on television
Reliability and validity of television
10 food pairs to form a maximum healthier preference score of 10. Similarly, knowledge of healthier foods was assessed by summing up the total correctly identified healthier food types out of the 10 food pairs to form a maximum knowledge score of 10. Section H consisted of six food types which were snack foods, sweets and chocolates, biscuits, fast food, breakfast cereals and carbonated drinks. These foods were considered as commonly advertised food in the free-to-air Malaysian TV channels (Karupaiah et al., 2008). All the food groups were similar to those listed by Stansfeld et al. (Stansfeld et al., 2003), except for ‘fried food, chips, samosas or bhajis, or fried English breakfast’ which has been replaced with fast foods. Breakfast cereal was added to the questionnaire as the sixth food group. For each food type, the participants were asked to record the frequency that they eat the food, whether more than once a day, once a day, at least once a week, once or twice a month or never. Scores were given to the frequencies. More than once a day was given one score; once a day, two scores; at least once a week, three scores; once or twice a month, four scores and never, five scores. Translation procedure The questionnaire was first drafted in English and later being translated to the national language, Bahasa Malaysia. The translation was checked and verified by the National Institute of Translation Malaysia. Data collection The study was conducted in April 2012 with permission obtained from the Ministry of Education Malaysia and the Selangor State Education Department. The questionnaire was administered twice to the same participants with 2-weeks interval to enable test –retest analysis. It was conducted in the classroom, whereby a briefing session was done by the researcher to all participants to ensure they understood what was required to be done. The participants had to answer the questions by themselves. However, they could always ask the researcher who was present all the time to clarify any item that they were not sure about. Quality check was done after all the questionnaires were collected while the participants were still present.
Downloaded from http://heapro.oxfordjournals.org/ at Western Michigan University on March 25, 2016
occupation, level of education and household income (Section A); the children’s date of birth, gender and race (Section B); the children’s weight, height and body mass index (Section C) and the time the children spent watching television (Section D). They were not included in the analysis to assess reliability and validity of the questionnaire. There were eight items in Section E. Each item had two, three or four answer options (Table 1). Each correct response was allocated one score and each incorrect response was allocated zero score. The maximum score for this section was eight. Sections F and G both consisted of the same 10 matched pairs of food types, one being healthier and the other less healthy. In Section F, for each food pair, the children were required to record which food type that they preferred and for Section G, the children were required to record which food type that they believed to be healthier. Even though the items were adapted Gwozdz and Reisch (Gwozdz and Reisch, 2011), some of the foods were modified or added to suit the local situation as recommended by the local experts. The 10 matched food pairs (in order of sequence) were sugar-free breakfast cereals versus sugared breakfast cereals, white rice versus fried rice (local food added), snack foods versus biscuits (local food added), carbonated drinks (modified from ‘Coke’) versus plain water, noodles prepared at home (modified from ‘pasta’) versus instant noodles (modified form ‘pot noodles’), fried chicken (modified from ‘beef burger’) versus chicken soup (modified from ‘roast beef’), fruit yoghurt (modified from ‘strawberry yoghurt’) versus fruit cake (modified from ‘strawberry cake’), wholemeal bread and white bread, fried potatoes (French fries) versus boiled potatoes (modified from potato) and fruits (modified from ‘orange’) versus fruit-flavoured ice-cream or ice confection (modified from ‘orange ice’. The following food pairs were not included in our questionnaire—cereal bar versus chocolate bar and orange juice versus orange squash. Food pairs were arranged side by side on a page. Photographs taken from local foods were placed beside the food item to help the participants to recognize the products. They were asked to put a smile on a face image next to the picture for the food they preferred (Section F) or believed to be healthier (Section G). Food preferences were assessed by summing up the total number of healthier (correct) food types preferred out of the
525
526
A. R. Zalma et al.
Table 1: Intra-rater reliability test for knowledge on television food advertising, knowledge on healthier food and preference for healthier food Dimensions
Items
Measurement of agreement (kappa value)
Knowledge on television food advertising (Section E)
(1) The main goal of an advertisement or commercial is to. . . (a) Entertain (incorrect) (b) Sell (correct) (c) Teach (incorrect) (d) Make you laugh (incorrect) (2) Photograph always shows people and things just the way they are in real life: (a) True (incorrect) (b) False (correct) (3) Food advertisements look fun, with bright colours and music, because. . . (a) The food is fun to eat (incorrect) (b) Parents like the ads (incorrect) (c) Fun ads get my attention (correct) (d) Watching food ads will make you a fun person (incorrect) (4) It is important to ask questions about what advertising tells us because. . . (a) Advertising makes us do things (incorrect) (b) Advertising is bad (incorrect) (c) Asking questions helps us make better choice (correct) (d) The teacher told me (incorrect) (5) Which is the best question to ask after seeing a commercial message that advertise candy or snack food? (a) Where can I buy this candy as soon as possible? (incorrect) (b) When will I be able to eat this candy? (incorrect) (c) Why was this message sent? (correct) (6) Everyone of my age likes the same advertisements on sweets that I like: True or false. (a) True (incorrect) (b) False (correct) (7) Television programmes might seem to be free but who ultimately pays for them? (a) Consumers (correct) (b) TV networks (incorrect) (c) Advertisers (incorrect) (8) It is important to consider who created the advertising message in order to. . . (a) Know who to blame (incorrect) (b) Find the bias that always exists (correct) (c) Find out who created the music on the commercial (incorrect) 1. Sugar-free breakfast cereals versus sugared breakfast cereals 2. White rice versus fried rice 3. Snack foods versus biscuits 4. Carbonated drinks versus plain water 5. Noodles prepared at home versus instant noodles 6. Fried chicken versus chicken soup 7. Fruit yoghurt versus fruit cake 8. Wholemeal bread and white bread 9. Fried potatoes (French fries) versus boiled potatoes 10. Fruits versus fruit-flavoured ice-cream or ice confection 1. Sugar-free breakfast cereals versus sugared breakfast cereals 2. White rice versus fried rice 3. Snack foods versus biscuits 4. Carbonated drinks versus plain water 5. Noodles prepared at home versus instant noodles 6. Fried chicken versus chicken soup 7. Fruit yoghurt versus fruit cake 8. Wholemeal bread and white bread 9. Fried potatoes (French fries) versus boiled potatoes 10. Fruits versus fruit-flavoured ice-cream or ice confection
0.558
Preference on healthier food (Section G)
0.553
0.403
0.793
0.262
0.737
0.209
0.599 0.783 0.611 0.413 0.793 0.843 0.757 0.512 0.923 0.712 0.475 0.533 0.625 0.64 0.68 0.784 0.713 0.331 0.587 0.585
Downloaded from http://heapro.oxfordjournals.org/ at Western Michigan University on March 25, 2016
Knowledge on healthier food (Section F)
0.595
Reliability and validity of television
Statistical analysis The mean scores for each of the sections (E, F, G and H) were calculated to analyse test– retest correlation to assess reliability. Kappa coefficient to assess intra-rater reliability was obtained using scores for each item under Sections E, F and G. Internal consistency and construct validity was assessed using Cronbach’s alpha coefficient and factor analysis, respectively, for six items from Section H. Significant level was set at 0.05 (2-tailed). All data were analysed using PASW Statistics 18 (SPSS, Inc., 2012).
All the 32 participants were Malays, 34.4% were male and 65.5% were female. The mean age was 129.5 + 3.7 months (10.8 years). The average time they spent watching television was 2 h per day. Thirty-one participants took part in the retest. Knowledge on television food advertising, knowledge on healthier food and preference for healthier food The results showed that the participants had the highest score for knowledge on healthier food, followed by preference for healthier food and knowledge on television food advertising for both test and retest. The mean score for test and retest for knowledge on healthier food was 7.69 +2.28 and 6.95 + 2.48, preference for healthier food was 5.13 + 2.03 and 4.93 + 1.91, and, knowledge on television food advertising was 3.63 + 1.58 and 3.65 + 1.66, respectively.
fast foods which had no significant correlation between test and retest scores. Kappa coefficient indicated acceptable-tostrong intra-rater reliability (K . 0.40), except for two items under knowledge on television food advertising and one item under preference for healthier food (Table 1). Internal consistency as showed by Cronbach’s alpha coefficient was low (0.573) when all the items under consumption of commonly advertised foods were analysed. The value became acceptable (0.610) after an item ‘consumption of breakfast cereal’ was deleted. However, the value of ‘Cronbach’s alpha if item deleted’ for ‘consumption of biscuits’ was higher (0.612) than this corresponding Cronbach’s alpha coefficient. Therefore, ‘consumption of biscuits’ was also deleted. After deleting these two items, Cronbach’s alpha coefficient showed acceptable internal consistency for the scores (Table 2). Even though the values were ,0.7, according to Di Iorio (Di Iorio, 2005), 0.7 should not be the only standard used to assess reliability, whereby the shorter scale, with the lower alpha value, actually demonstrates higher interrelatedness among items. Furthermore, the total corrected correlation of each item was .0.2, and the value of ‘Cronbach’s alpha if item deleted’ for each item was less than the corresponding Cronbach’s alpha coefficient of the dimension, showing that there should be no item to be eliminated or revised. Validity of the questionnaire Factor analysis was used to assess the validity of the questionnaire. However, the analysis was only Table 2: Internal consistency for consumption of commonly advertised foods
Reliability of the questionnaire The test–retest correlation coefficient, Spearman’s rho for knowledge on television food advertising was 0.690 ( p , 0.01), knowledge on healthier food was 0.850 ( p , 0.01) and preference for healthier food was 0.950 ( p , 0.01). These indicated moderate-to-high reliability for all scores. For consumption of commonly advertised foods, the test–retest correlation coefficient, Spearman’s rho was 0.595 (p , 0.01) for snack foods, 0.533 (p , 0.01) for sweets and chocolates, 0.648 (p , 0.01) for biscuits, 0.238 (p . 0.05) for fast foods, 0.410 (p , 0.05) for breakfast cereals and 0.405 (p , 0.05) for carbonated drinks. These indicated moderate reliability, except for consumption of
Items
Corrected item-total correlation
Cronbach’s alpha if item deleted
Cronbach’s alpha
Consumption of snack foods Consumption of sweets and chocolates Consumption of fast foods Consumption of carbonated drinks
0.337
0.588
0.612
0.548
0.451
0.407
0.530
0.318
0.598
Downloaded from http://heapro.oxfordjournals.org/ at Western Michigan University on March 25, 2016
RESULTS
527
528
A. R. Zalma et al.
done for Part H which had items with more than three response options. One basic requirement for factor analysis is that the scale includes items that are measured on a rating scale that consists of at least three response options (Di Iorio, 2005). After deleting two items under consumption of commonly advertised food, the items showed moderate-to-high loading (Table 3), with the Scree plot showing that there was only one factor. The KMO (Kaiser –Meyer –Olkin) was 0.60, showing that the sample was adequate for factor analysis (Field, 2005).
This study focused on developing an instrument to assess the effect of media literacy education on television food advertising on schoolchildren. The effect was assessed in four dimensions, which were knowledge on television food advertising, knowledge on healthier food, preference for healthier food and consumption of commonly advertised food. The average time the children spent watching television was 2 h per day. This shows that this group of children was appropriate sample for this study, whereby children who watched TV two or more hours a day was more likely to be consumers of all the commonly advertised foods (Utter et al., 2006). The children had the highest score for knowledge on healthier food. This may be due to fact that nutrition is one of the health component taught in schools through various subjects such as Science, Physical Education and Health, Living Skills, Bahasa Malaysia and English. However, scores for preference for healthier food was not as high as knowledge on healthier foods, showing that the children do not necessarily choose or eat healthier food even though they know that the food is healthier. This finding shows that lack of nutrition knowledge may not Table 3: Factor analysis for consumption of commonly advertised foods Items
Factor loading (n ¼ 32)
Consumption of snack foods Consumption of sweets and chocolates Consumption of fast foods Consumption of carbonated drinks
0.523 0.835 0.423 0.417
Downloaded from http://heapro.oxfordjournals.org/ at Western Michigan University on March 25, 2016
DISCUSSION
be the reason for poor eating habits among schoolchildren in Malaysia. It is consistent with previous findings among UK children (Kopelman et al., 2007; Gwozdz and Reisch, 2011). The scores for knowledge on healthier food and preference for healthier food was lower at the retest. The participants might had been bored and less motivated since they had to answer the questions twice. This fatigue effect leads to inaccurate responses by the children. The same trend was reported in another study by Vereecken et al. (Vereecken et al., 2012). However, this seems not to give significant effect to the stability of the questionnaire as the test– retest correlation indicates moderate-to-high reliability for all scores. Strong intra-reliability of the questionnaire showed that it is stable and reproducible. However, low Kappa value for two items under knowledge on television food advertising showed that the items need to be reviewed to make it better understood by the children. The same goes to the item under preference for healthier foods (wholemeal bread and white bread). The children may not be exposed to or consume different types of bread; therefore, they were not being able to choose constantly the bread that they prefer. This result suggests that explanation on the types of bread need to be thorough and in-depth discussion on the food types that the children are consuming is crucial before listing the food types that they prefer. Two items under consumptions of commonly advertised foods which were consumptions of biscuits and consumptions of breakfast cereals were deleted to have the good internal consistency among the items in this dimension. The foods listed under this dimension were considered as less healthy because most food products advertised on television are high in sugar, fat and salt (WHO, 2006). However, for these children, they believe that biscuits and breakfast cereals are healthy due to repeated messages on television portraying all biscuit and breakfast cereals are healthy. The fact is that most television food advertisements were for sugary biscuits and breakfast cereals. Biscuit was one of the food groups that have the highest energy density and the highest fat content in the advertised foods (Karupaiah et al., 2008). This result suggests the need to include examples of these foods (less healthy versions that are commonly advertised on television) in the questionnaire. Factor analysis showed that the questions under consumptions of commonly advertised foods were valid. It was loaded with one factor
Reliability and validity of television
ACKNOWLEDGEMENTS The authors acknowledge contributions from the Ministry of Education Malaysia, the Selangor State Education Department, teachers and students who enthusiastically participated in this study. FUNDING This work was supported by a grant from Universiti Teknologi MARA (UiTM) (600-RMI/ DANA 5/3/RIF(331/2012).
REFERENCES Brown, J. D. (2001) Statistics Corner: Questions and Answers about Language Testing Statistics: What is Eigenvalue? JALT Testing & Evaluation SIG Newsletter, Shiken. Cairns, G., Angus, K. and Hastings, G. (2009) The Extent, Nature and Effects of Food Promotion to Children: A Review of the Evidence to December 2008. World Health Organization, Geneva, Switzerland. Di Iorio, C. K. (2005) Measurement in health behaviour: methods for research evaluation. Chapter 10. JosseyBass, San Francisco, USA, pp. 176– 210. Field, A. (2005). Factor analysis using SPSS. http://www.statisticshell.com/factor.pdf (last assessed 2 September 2012). Gwozdz, W. and Reisch, L. A. (2011) Instruments for analysing the influence of advertising ‘on children’s food choices. International Journal of Obesity, 35, S137–S143. Harris, J. L. and Bargh, J. A. (2009) Television viewing and unhealthy diet: implications for children and media interventions. Health Communication, 24, 660–673. Hassan, O. R. and Rasiah, R. (2011) Poverty and student performance in Malaysia. International Journal of Institutions and Economics, 3, 61– 76. Hawkes, C. (2004) Marketing Food to Children: The Global Regulatory Environment. World Health Organization, Geneva, Switzerland. Jolls, T. (2011) Educator Guide. A Recipe for Action: Deconstructing Food Advertising. Center for Media Literacy, USA. Karupaiah, T., Chinna, K., Mee, L. H., Mei, l. S. and Noor, M. I. (2008) What’s on Malaysian television?—a survey on food advertising targeting children. Asia Pacific Journal of Clinical Nutrition, 17, 483–491. Kelly, B., Halford, J. C. G., Boyland, E. J., Chapman, K., Bautista-Castan˜o, I., Berg, C. et al. (2010) Television food advertising to children: a global perspective. American Journal of Public Health, 100, 1730– 1736. Kopelman, C. A., Roberts, L. M. and Adab, P. (2007) Advertising of food to children: is brand logo recognition related to their food knowledge, eating behaviours and food preferences. Journal of Public Health, 29, 358– 367 (http://jpubhealth.oxfordjournals.org/ (last accessed 12 December 2011). Livingston, S. and Helpers, E. (2004) Advertising Foods to Children: Understanding Promotion in the Context of Children’s Daily Lives. Dept of Media and Communications, London School of Economics and Political Science, London, UK. MOH. (2008) The Third National Health and Morbidity Survey 2006 (NHMS III): Nutritional Status. Chapter 1. Institute for Public Health, Ministry of Health Malaysia, Kuala Lumpur, pp. 3 –44. Nazari, M. R., Hassan, M. S. B. H., Parhizkar, S. and Hassan, M. B. A. (2011) Correlation between children’s television advertising exposure and their food preference. Journal of Media and Communication Studies, 3, 263–268 (http://www.academicjournals.org/JMCS (last assessed 29 May 2012). Pett, M. A., Lackey, N. R. and Sullivan, J. J. (2003) Making Sense of Factor Analysis: The Use of Factor Analysis for Instrument Development in Health Care Research. Sage, Thousand Oaks, CA. Stansfeld, S., Haines, M., Booy, R., Taylor, S., Viner, R., Head, J. et al. (2003) Health of young people in East
Downloaded from http://heapro.oxfordjournals.org/ at Western Michigan University on March 25, 2016
which was shown by the Scree plot, the plot that is being used to determine the number of factors that are appropriate with the instrument (Brown, 2001). It proved that all the items were asking the same factor, which is less healthy food which was commonly advertised on television. Even though the sample is small, the value of KMO was 0.6, showing that it is adequate for factor analysis, as the value of KMO should be .0.5 (Field, 2005). Furthermore, the sample of 32 for four items is not really far below the recommendation by Pett et al. (Pett et al., 2003) who suggest 10–15 participants per item. The limitation of the study is that it was conducted in the urban area and the participants were all Malays. For participants in rural areas, a simpler questionnaire might be needed since their level of understanding of the questions might be lower. Children from rural areas generally have lower school achievement (Hassan and Rasiah, 2011). For children from other races especially who are attending national-type schools which use Mandarin or Tamil as the medium, they might not well understood the questions set in Bahasa Malaysia. A questionnaire set in Mandarin and Tamil might be needed for these children. Therefore, other studies need to be conducted to determine reliability and validity of this questionnaire amongst rural children and also children from different ethnic groups. Since this study has a small sample size and only construct validity was assessed, other studies with bigger sample size and aimed at assessing other forms of validity (i.e. discriminative and predictive validity) need to be conducted too. Overall, the questionnaire on television food advertising is reliable and valid to assess the effect of media literacy education on television food advertising on schoolchildren.
529
530
A. R. Zalma et al.
London: The RELACHS Study 2001. The Stationery Office, London, UK. Utter, J., Scragg, R. and Schaaf, D. (2006) Associations between television viewing and consumption of commonly advertised foods among New Zealand children and young adolescents. Public Health Nutrition, 9, 606– 612. http:www.journals.cambridge.org (last accessed 14 December 2011). Vereecken, C., De Pauw, A., Van Cauwenbergh, S. and Maes, L. (2012) Development and test–retest reliability
of a nutrition knowledge questionnaire for primaryschool children. Public Health Nutrition, 15, 1630–1638. WHO. (2006) Marketing of food and non-alcoholic beverages to children. Report of a WHO Forum and Technical Meeting, Oslo, Norway, 2 –5 May 2006. World Health Organization, Geneva, Switzerland. WHO. (2012) A Framework for Implementing the Set of Recommendations on the Marketing of Foods and Non-Alcoholic Beverages to Children. World Health Organization, Geneva, Switzerland.
Downloaded from http://heapro.oxfordjournals.org/ at Western Michigan University on March 25, 2016
# The Author (2013). Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
Reliability and validity of television food advertising questionnaire in Malaysia 1
Faculty of Health Sciences, Universiti Teknologi MARA, Puncak Alam Campus, 42300 Puncak Alam, Selangor, Malaysia and 2Nutrition Division, Ministry of Health Malaysia, Level 1, Block E3, Parcel E, Precint 1, Federal Government Administrative Center, 62590 Putrajaya, Malaysia *Corresponding author. E-mail: [email protected]
SUMMARY Interventions to counter the influence of television food advertising amongst children are important. Thus, reliable and valid instrument to assess its effect is needed. The objective of this study was to determine the reliability and validity of such a questionnaire. The questionnaire was administered twice on 32 primary schoolchildren aged 10– 11 years in Selangor, Malaysia. The interval between the first and second administration was 2 weeks. Test– retest method was used to examine the reliability of the questionnaire. Intra-rater reliability was determined by kappa coefficient and internal consistency by Cronbach’s alpha coefficient. Construct validity was evaluated using factor analysis. The test– retest correlation showed moderate-tohigh reliability for all scores (r ¼ 0.40*, p ¼ 0.02 to r ¼ 0.95**, p ¼ 0.00), with one exception, consumption of fast
foods (r ¼ 0.24, p ¼ 0.20). Kappa coefficient showed acceptable-to-strong intra-rater reliability (K ¼ 0.40– 0.92), except for two items under knowledge on television food advertising (K ¼ 0.26 and K ¼ 0.21) and one item under preference for healthier foods (K ¼ 0.33). Cronbach’s alpha coefficient indicated acceptable internal consistency for all scores (0.45–0.60). After deleting two items under Consumption of Commonly Advertised Food, the items showed moderate-to-high loading (0.52, 0.84, 0.42 and 0.42) with the Scree plot showing that there was only one factor. The Kaiser –Meyer –Olkin was 0.60, showing that the sample was adequate for factor analysis. The questionnaire on television food advertising is reliable and valid to assess the effect of media literacy education on television food advertising on schoolchildren.
Key words: reliability; validity; food advertising; schoolchildren
INTRODUCTION Childhood obesity is a global problem and it is a public health concern in Malaysia. The National Health and Morbidity Survey (NHMS) III reported that 5.4% of children aged ,18 years were overweight (weight-for-age . þ2SD) (MOH, 2008), whereas the reanalysed data of the NHMS III showed that 11.4% children aged 6–12 years were obese (BMI-for-age . þ2SD) (MOH, unpublished data). Whilst many factors contribute to obesity, food advertisements targeted at children has been recognized as one of the main factors, and
it is considered as an important arena of action in the prevention of obesity (WHO, 2006). Food industries are using various strategies to get to the children, and the most dominant medium used is television (Cairns et al., 2009). This is because television is the most effective way of marketing foods worldwide (Hawkes, 2004). Studies done by 13 independent research groups in 11 countries found that 18% of the television advertisements were for food, and overall, food was the second most frequently advertised product, after channel promotions (23%). Majority of food advertisements were for 523
Downloaded from http://heapro.oxfordjournals.org/ at Western Michigan University on March 25, 2016
ABDUL RAZAK ZALMA1,2*, MD. YUSOF SAFIAH1, DANIS AJAU1 and MD. ISA KHAIRIL ANUAR1
524
A. R. Zalma et al.
food advertising, which will be used to assess the effect of media literacy education on television food advertising among schoolchildren. Since food advertising influences children’s nutritional knowledge, food preferences and food consumptions (WHO, 2006), the questionnaire was developed to cover these three main dimensions which were knowledge on healthier food, preference for healthier food and consumption of commonly advertised foods on television. Knowledge on television food advertising was also included to gather some basic data on the children’s media literacy. METHODS This was a pilot study of a questionnaire intended to be used in a study to evaluate the effect of media literacy education on television food advertising on schoolchildren. The study protocol was approved by the Ethical Committee of Universiti Teknologi MARA Malaysia. Sample For this study a primary school in Selangor (one of the states in Malaysia) was selected and the participants were all the pupils from one class that had been selected at random. At the first administration of the questionnaire, 32 participants aged 10–11 years took part, and at the second administration 31 took part. Television food advertising questionnaire The questionnaire was developed to cover the three areas that can be influenced by food promotion to children (WHO, 2006). These are children’s nutritional knowledge, food preferences and food consumption. The other areas, which were purchasing and purchase-related behaviour and diet and health status, were not covered in this questionnaire. Items for knowledge on television food advertising (section E) were adapted from Jolls (Jolls, 2011), whereas items for knowledge on healthier food (section F) and preference for healthier food (section G) were adapted from Gwozdz and Reisch (Gwozdz and Reisch, 2011). Items for Consumption of Commonly Advertised Foods (section H) were adapted from Stansfeld et al. (Stansfeld et al., 2003). Sections A –D gather basic information of the children and their parents which are the parents’
Downloaded from http://heapro.oxfordjournals.org/ at Western Michigan University on March 25, 2016
non-core (unhealthy) food (67%), such as fast-food restaurant meals (12% of food advertisements), chocolate and confectionery (12%) and high-fat, high-sugar or high-salt spreads and sauces (8%) (Kelly et al., 2010). In a study conducted in Malaysia, it was found that 54162 advertisements were broadcasted by free-to-air TV stations during children’s prime time, and 27.9% composed of food products advertisements (Karupaiah et al., 2008). Snack foods which are considered as unhealthy food was found to be the most advertised food groups (34.5%), followed by dairy products (20.3%), confectionary (13.4%), biscuits (11.2%), fast food (6.7%), breakfast cereals (6.4%) and beverages (4.1%). Children need to be protected from heavy marketing of unhealthy food through television food advertising because marketing generates positive beliefs about advertised foods and influences children’s food preferences, purchase requests and consumption (Livingston and Helpers, 2004; Cairns et al., 2009; Nazari et al., 2011). The influence of food marketing is beyond their understanding of its persuasive intent (WHO, 2012). Harris and Bargh (Harris and Bargh, 2009) had proposed media literacy education as one of the three broad categories of solutions to protect children against the unhealthy influence of television and food advertising. Media literacy means the ability to gain access, analyse, evaluate, create and participate with media in all its forms (Jolls, 2011), thus media literacy education means teaching the skills associated with media literacy. Critical viewing skills acquired through media literacy education has been found to be related to higher taste ratings for healthy foods and lower taste ratings for unhealthy, highly advertised foods (Harris and Bargh, 2009). Thus, teaching children to defend against advertising influence through media literacy education is recommended to increase their critical viewing skills and scepticism about the media and advertising. Media literacy education on television food advertising is relatively new in Malaysia and to our knowledge there is no standard instrument developed for our population to determine the effectiveness of media literacy education. Other studies (Stansfeld et al., 2003; Gwozdz and Reisch, 2011; Jolls, 2011) regarding food advertising and food consumption have used instruments suitable for European and American population. The objective of this study was to examine the reliability and validity of a questionnaire on television
Reliability and validity of television
10 food pairs to form a maximum healthier preference score of 10. Similarly, knowledge of healthier foods was assessed by summing up the total correctly identified healthier food types out of the 10 food pairs to form a maximum knowledge score of 10. Section H consisted of six food types which were snack foods, sweets and chocolates, biscuits, fast food, breakfast cereals and carbonated drinks. These foods were considered as commonly advertised food in the free-to-air Malaysian TV channels (Karupaiah et al., 2008). All the food groups were similar to those listed by Stansfeld et al. (Stansfeld et al., 2003), except for ‘fried food, chips, samosas or bhajis, or fried English breakfast’ which has been replaced with fast foods. Breakfast cereal was added to the questionnaire as the sixth food group. For each food type, the participants were asked to record the frequency that they eat the food, whether more than once a day, once a day, at least once a week, once or twice a month or never. Scores were given to the frequencies. More than once a day was given one score; once a day, two scores; at least once a week, three scores; once or twice a month, four scores and never, five scores. Translation procedure The questionnaire was first drafted in English and later being translated to the national language, Bahasa Malaysia. The translation was checked and verified by the National Institute of Translation Malaysia. Data collection The study was conducted in April 2012 with permission obtained from the Ministry of Education Malaysia and the Selangor State Education Department. The questionnaire was administered twice to the same participants with 2-weeks interval to enable test –retest analysis. It was conducted in the classroom, whereby a briefing session was done by the researcher to all participants to ensure they understood what was required to be done. The participants had to answer the questions by themselves. However, they could always ask the researcher who was present all the time to clarify any item that they were not sure about. Quality check was done after all the questionnaires were collected while the participants were still present.
Downloaded from http://heapro.oxfordjournals.org/ at Western Michigan University on March 25, 2016
occupation, level of education and household income (Section A); the children’s date of birth, gender and race (Section B); the children’s weight, height and body mass index (Section C) and the time the children spent watching television (Section D). They were not included in the analysis to assess reliability and validity of the questionnaire. There were eight items in Section E. Each item had two, three or four answer options (Table 1). Each correct response was allocated one score and each incorrect response was allocated zero score. The maximum score for this section was eight. Sections F and G both consisted of the same 10 matched pairs of food types, one being healthier and the other less healthy. In Section F, for each food pair, the children were required to record which food type that they preferred and for Section G, the children were required to record which food type that they believed to be healthier. Even though the items were adapted Gwozdz and Reisch (Gwozdz and Reisch, 2011), some of the foods were modified or added to suit the local situation as recommended by the local experts. The 10 matched food pairs (in order of sequence) were sugar-free breakfast cereals versus sugared breakfast cereals, white rice versus fried rice (local food added), snack foods versus biscuits (local food added), carbonated drinks (modified from ‘Coke’) versus plain water, noodles prepared at home (modified from ‘pasta’) versus instant noodles (modified form ‘pot noodles’), fried chicken (modified from ‘beef burger’) versus chicken soup (modified from ‘roast beef’), fruit yoghurt (modified from ‘strawberry yoghurt’) versus fruit cake (modified from ‘strawberry cake’), wholemeal bread and white bread, fried potatoes (French fries) versus boiled potatoes (modified from potato) and fruits (modified from ‘orange’) versus fruit-flavoured ice-cream or ice confection (modified from ‘orange ice’. The following food pairs were not included in our questionnaire—cereal bar versus chocolate bar and orange juice versus orange squash. Food pairs were arranged side by side on a page. Photographs taken from local foods were placed beside the food item to help the participants to recognize the products. They were asked to put a smile on a face image next to the picture for the food they preferred (Section F) or believed to be healthier (Section G). Food preferences were assessed by summing up the total number of healthier (correct) food types preferred out of the
525
526
A. R. Zalma et al.
Table 1: Intra-rater reliability test for knowledge on television food advertising, knowledge on healthier food and preference for healthier food Dimensions
Items
Measurement of agreement (kappa value)
Knowledge on television food advertising (Section E)
(1) The main goal of an advertisement or commercial is to. . . (a) Entertain (incorrect) (b) Sell (correct) (c) Teach (incorrect) (d) Make you laugh (incorrect) (2) Photograph always shows people and things just the way they are in real life: (a) True (incorrect) (b) False (correct) (3) Food advertisements look fun, with bright colours and music, because. . . (a) The food is fun to eat (incorrect) (b) Parents like the ads (incorrect) (c) Fun ads get my attention (correct) (d) Watching food ads will make you a fun person (incorrect) (4) It is important to ask questions about what advertising tells us because. . . (a) Advertising makes us do things (incorrect) (b) Advertising is bad (incorrect) (c) Asking questions helps us make better choice (correct) (d) The teacher told me (incorrect) (5) Which is the best question to ask after seeing a commercial message that advertise candy or snack food? (a) Where can I buy this candy as soon as possible? (incorrect) (b) When will I be able to eat this candy? (incorrect) (c) Why was this message sent? (correct) (6) Everyone of my age likes the same advertisements on sweets that I like: True or false. (a) True (incorrect) (b) False (correct) (7) Television programmes might seem to be free but who ultimately pays for them? (a) Consumers (correct) (b) TV networks (incorrect) (c) Advertisers (incorrect) (8) It is important to consider who created the advertising message in order to. . . (a) Know who to blame (incorrect) (b) Find the bias that always exists (correct) (c) Find out who created the music on the commercial (incorrect) 1. Sugar-free breakfast cereals versus sugared breakfast cereals 2. White rice versus fried rice 3. Snack foods versus biscuits 4. Carbonated drinks versus plain water 5. Noodles prepared at home versus instant noodles 6. Fried chicken versus chicken soup 7. Fruit yoghurt versus fruit cake 8. Wholemeal bread and white bread 9. Fried potatoes (French fries) versus boiled potatoes 10. Fruits versus fruit-flavoured ice-cream or ice confection 1. Sugar-free breakfast cereals versus sugared breakfast cereals 2. White rice versus fried rice 3. Snack foods versus biscuits 4. Carbonated drinks versus plain water 5. Noodles prepared at home versus instant noodles 6. Fried chicken versus chicken soup 7. Fruit yoghurt versus fruit cake 8. Wholemeal bread and white bread 9. Fried potatoes (French fries) versus boiled potatoes 10. Fruits versus fruit-flavoured ice-cream or ice confection
0.558
Preference on healthier food (Section G)
0.553
0.403
0.793
0.262
0.737
0.209
0.599 0.783 0.611 0.413 0.793 0.843 0.757 0.512 0.923 0.712 0.475 0.533 0.625 0.64 0.68 0.784 0.713 0.331 0.587 0.585
Downloaded from http://heapro.oxfordjournals.org/ at Western Michigan University on March 25, 2016
Knowledge on healthier food (Section F)
0.595
Reliability and validity of television
Statistical analysis The mean scores for each of the sections (E, F, G and H) were calculated to analyse test– retest correlation to assess reliability. Kappa coefficient to assess intra-rater reliability was obtained using scores for each item under Sections E, F and G. Internal consistency and construct validity was assessed using Cronbach’s alpha coefficient and factor analysis, respectively, for six items from Section H. Significant level was set at 0.05 (2-tailed). All data were analysed using PASW Statistics 18 (SPSS, Inc., 2012).
All the 32 participants were Malays, 34.4% were male and 65.5% were female. The mean age was 129.5 + 3.7 months (10.8 years). The average time they spent watching television was 2 h per day. Thirty-one participants took part in the retest. Knowledge on television food advertising, knowledge on healthier food and preference for healthier food The results showed that the participants had the highest score for knowledge on healthier food, followed by preference for healthier food and knowledge on television food advertising for both test and retest. The mean score for test and retest for knowledge on healthier food was 7.69 +2.28 and 6.95 + 2.48, preference for healthier food was 5.13 + 2.03 and 4.93 + 1.91, and, knowledge on television food advertising was 3.63 + 1.58 and 3.65 + 1.66, respectively.
fast foods which had no significant correlation between test and retest scores. Kappa coefficient indicated acceptable-tostrong intra-rater reliability (K . 0.40), except for two items under knowledge on television food advertising and one item under preference for healthier food (Table 1). Internal consistency as showed by Cronbach’s alpha coefficient was low (0.573) when all the items under consumption of commonly advertised foods were analysed. The value became acceptable (0.610) after an item ‘consumption of breakfast cereal’ was deleted. However, the value of ‘Cronbach’s alpha if item deleted’ for ‘consumption of biscuits’ was higher (0.612) than this corresponding Cronbach’s alpha coefficient. Therefore, ‘consumption of biscuits’ was also deleted. After deleting these two items, Cronbach’s alpha coefficient showed acceptable internal consistency for the scores (Table 2). Even though the values were ,0.7, according to Di Iorio (Di Iorio, 2005), 0.7 should not be the only standard used to assess reliability, whereby the shorter scale, with the lower alpha value, actually demonstrates higher interrelatedness among items. Furthermore, the total corrected correlation of each item was .0.2, and the value of ‘Cronbach’s alpha if item deleted’ for each item was less than the corresponding Cronbach’s alpha coefficient of the dimension, showing that there should be no item to be eliminated or revised. Validity of the questionnaire Factor analysis was used to assess the validity of the questionnaire. However, the analysis was only Table 2: Internal consistency for consumption of commonly advertised foods
Reliability of the questionnaire The test–retest correlation coefficient, Spearman’s rho for knowledge on television food advertising was 0.690 ( p , 0.01), knowledge on healthier food was 0.850 ( p , 0.01) and preference for healthier food was 0.950 ( p , 0.01). These indicated moderate-to-high reliability for all scores. For consumption of commonly advertised foods, the test–retest correlation coefficient, Spearman’s rho was 0.595 (p , 0.01) for snack foods, 0.533 (p , 0.01) for sweets and chocolates, 0.648 (p , 0.01) for biscuits, 0.238 (p . 0.05) for fast foods, 0.410 (p , 0.05) for breakfast cereals and 0.405 (p , 0.05) for carbonated drinks. These indicated moderate reliability, except for consumption of
Items
Corrected item-total correlation
Cronbach’s alpha if item deleted
Cronbach’s alpha
Consumption of snack foods Consumption of sweets and chocolates Consumption of fast foods Consumption of carbonated drinks
0.337
0.588
0.612
0.548
0.451
0.407
0.530
0.318
0.598
Downloaded from http://heapro.oxfordjournals.org/ at Western Michigan University on March 25, 2016
RESULTS
527
528
A. R. Zalma et al.
done for Part H which had items with more than three response options. One basic requirement for factor analysis is that the scale includes items that are measured on a rating scale that consists of at least three response options (Di Iorio, 2005). After deleting two items under consumption of commonly advertised food, the items showed moderate-to-high loading (Table 3), with the Scree plot showing that there was only one factor. The KMO (Kaiser –Meyer –Olkin) was 0.60, showing that the sample was adequate for factor analysis (Field, 2005).
This study focused on developing an instrument to assess the effect of media literacy education on television food advertising on schoolchildren. The effect was assessed in four dimensions, which were knowledge on television food advertising, knowledge on healthier food, preference for healthier food and consumption of commonly advertised food. The average time the children spent watching television was 2 h per day. This shows that this group of children was appropriate sample for this study, whereby children who watched TV two or more hours a day was more likely to be consumers of all the commonly advertised foods (Utter et al., 2006). The children had the highest score for knowledge on healthier food. This may be due to fact that nutrition is one of the health component taught in schools through various subjects such as Science, Physical Education and Health, Living Skills, Bahasa Malaysia and English. However, scores for preference for healthier food was not as high as knowledge on healthier foods, showing that the children do not necessarily choose or eat healthier food even though they know that the food is healthier. This finding shows that lack of nutrition knowledge may not Table 3: Factor analysis for consumption of commonly advertised foods Items
Factor loading (n ¼ 32)
Consumption of snack foods Consumption of sweets and chocolates Consumption of fast foods Consumption of carbonated drinks
0.523 0.835 0.423 0.417
Downloaded from http://heapro.oxfordjournals.org/ at Western Michigan University on March 25, 2016
DISCUSSION
be the reason for poor eating habits among schoolchildren in Malaysia. It is consistent with previous findings among UK children (Kopelman et al., 2007; Gwozdz and Reisch, 2011). The scores for knowledge on healthier food and preference for healthier food was lower at the retest. The participants might had been bored and less motivated since they had to answer the questions twice. This fatigue effect leads to inaccurate responses by the children. The same trend was reported in another study by Vereecken et al. (Vereecken et al., 2012). However, this seems not to give significant effect to the stability of the questionnaire as the test– retest correlation indicates moderate-to-high reliability for all scores. Strong intra-reliability of the questionnaire showed that it is stable and reproducible. However, low Kappa value for two items under knowledge on television food advertising showed that the items need to be reviewed to make it better understood by the children. The same goes to the item under preference for healthier foods (wholemeal bread and white bread). The children may not be exposed to or consume different types of bread; therefore, they were not being able to choose constantly the bread that they prefer. This result suggests that explanation on the types of bread need to be thorough and in-depth discussion on the food types that the children are consuming is crucial before listing the food types that they prefer. Two items under consumptions of commonly advertised foods which were consumptions of biscuits and consumptions of breakfast cereals were deleted to have the good internal consistency among the items in this dimension. The foods listed under this dimension were considered as less healthy because most food products advertised on television are high in sugar, fat and salt (WHO, 2006). However, for these children, they believe that biscuits and breakfast cereals are healthy due to repeated messages on television portraying all biscuit and breakfast cereals are healthy. The fact is that most television food advertisements were for sugary biscuits and breakfast cereals. Biscuit was one of the food groups that have the highest energy density and the highest fat content in the advertised foods (Karupaiah et al., 2008). This result suggests the need to include examples of these foods (less healthy versions that are commonly advertised on television) in the questionnaire. Factor analysis showed that the questions under consumptions of commonly advertised foods were valid. It was loaded with one factor
Reliability and validity of television
ACKNOWLEDGEMENTS The authors acknowledge contributions from the Ministry of Education Malaysia, the Selangor State Education Department, teachers and students who enthusiastically participated in this study. FUNDING This work was supported by a grant from Universiti Teknologi MARA (UiTM) (600-RMI/ DANA 5/3/RIF(331/2012).
REFERENCES Brown, J. D. (2001) Statistics Corner: Questions and Answers about Language Testing Statistics: What is Eigenvalue? JALT Testing & Evaluation SIG Newsletter, Shiken. Cairns, G., Angus, K. and Hastings, G. (2009) The Extent, Nature and Effects of Food Promotion to Children: A Review of the Evidence to December 2008. World Health Organization, Geneva, Switzerland. Di Iorio, C. K. (2005) Measurement in health behaviour: methods for research evaluation. Chapter 10. JosseyBass, San Francisco, USA, pp. 176– 210. Field, A. (2005). Factor analysis using SPSS. http://www.statisticshell.com/factor.pdf (last assessed 2 September 2012). Gwozdz, W. and Reisch, L. A. (2011) Instruments for analysing the influence of advertising ‘on children’s food choices. International Journal of Obesity, 35, S137–S143. Harris, J. L. and Bargh, J. A. (2009) Television viewing and unhealthy diet: implications for children and media interventions. Health Communication, 24, 660–673. Hassan, O. R. and Rasiah, R. (2011) Poverty and student performance in Malaysia. International Journal of Institutions and Economics, 3, 61– 76. Hawkes, C. (2004) Marketing Food to Children: The Global Regulatory Environment. World Health Organization, Geneva, Switzerland. Jolls, T. (2011) Educator Guide. A Recipe for Action: Deconstructing Food Advertising. Center for Media Literacy, USA. Karupaiah, T., Chinna, K., Mee, L. H., Mei, l. S. and Noor, M. I. (2008) What’s on Malaysian television?—a survey on food advertising targeting children. Asia Pacific Journal of Clinical Nutrition, 17, 483–491. Kelly, B., Halford, J. C. G., Boyland, E. J., Chapman, K., Bautista-Castan˜o, I., Berg, C. et al. (2010) Television food advertising to children: a global perspective. American Journal of Public Health, 100, 1730– 1736. Kopelman, C. A., Roberts, L. M. and Adab, P. (2007) Advertising of food to children: is brand logo recognition related to their food knowledge, eating behaviours and food preferences. Journal of Public Health, 29, 358– 367 (http://jpubhealth.oxfordjournals.org/ (last accessed 12 December 2011). Livingston, S. and Helpers, E. (2004) Advertising Foods to Children: Understanding Promotion in the Context of Children’s Daily Lives. Dept of Media and Communications, London School of Economics and Political Science, London, UK. MOH. (2008) The Third National Health and Morbidity Survey 2006 (NHMS III): Nutritional Status. Chapter 1. Institute for Public Health, Ministry of Health Malaysia, Kuala Lumpur, pp. 3 –44. Nazari, M. R., Hassan, M. S. B. H., Parhizkar, S. and Hassan, M. B. A. (2011) Correlation between children’s television advertising exposure and their food preference. Journal of Media and Communication Studies, 3, 263–268 (http://www.academicjournals.org/JMCS (last assessed 29 May 2012). Pett, M. A., Lackey, N. R. and Sullivan, J. J. (2003) Making Sense of Factor Analysis: The Use of Factor Analysis for Instrument Development in Health Care Research. Sage, Thousand Oaks, CA. Stansfeld, S., Haines, M., Booy, R., Taylor, S., Viner, R., Head, J. et al. (2003) Health of young people in East
Downloaded from http://heapro.oxfordjournals.org/ at Western Michigan University on March 25, 2016
which was shown by the Scree plot, the plot that is being used to determine the number of factors that are appropriate with the instrument (Brown, 2001). It proved that all the items were asking the same factor, which is less healthy food which was commonly advertised on television. Even though the sample is small, the value of KMO was 0.6, showing that it is adequate for factor analysis, as the value of KMO should be .0.5 (Field, 2005). Furthermore, the sample of 32 for four items is not really far below the recommendation by Pett et al. (Pett et al., 2003) who suggest 10–15 participants per item. The limitation of the study is that it was conducted in the urban area and the participants were all Malays. For participants in rural areas, a simpler questionnaire might be needed since their level of understanding of the questions might be lower. Children from rural areas generally have lower school achievement (Hassan and Rasiah, 2011). For children from other races especially who are attending national-type schools which use Mandarin or Tamil as the medium, they might not well understood the questions set in Bahasa Malaysia. A questionnaire set in Mandarin and Tamil might be needed for these children. Therefore, other studies need to be conducted to determine reliability and validity of this questionnaire amongst rural children and also children from different ethnic groups. Since this study has a small sample size and only construct validity was assessed, other studies with bigger sample size and aimed at assessing other forms of validity (i.e. discriminative and predictive validity) need to be conducted too. Overall, the questionnaire on television food advertising is reliable and valid to assess the effect of media literacy education on television food advertising on schoolchildren.
529
530
A. R. Zalma et al.
London: The RELACHS Study 2001. The Stationery Office, London, UK. Utter, J., Scragg, R. and Schaaf, D. (2006) Associations between television viewing and consumption of commonly advertised foods among New Zealand children and young adolescents. Public Health Nutrition, 9, 606– 612. http:www.journals.cambridge.org (last accessed 14 December 2011). Vereecken, C., De Pauw, A., Van Cauwenbergh, S. and Maes, L. (2012) Development and test–retest reliability
of a nutrition knowledge questionnaire for primaryschool children. Public Health Nutrition, 15, 1630–1638. WHO. (2006) Marketing of food and non-alcoholic beverages to children. Report of a WHO Forum and Technical Meeting, Oslo, Norway, 2 –5 May 2006. World Health Organization, Geneva, Switzerland. WHO. (2012) A Framework for Implementing the Set of Recommendations on the Marketing of Foods and Non-Alcoholic Beverages to Children. World Health Organization, Geneva, Switzerland.
Downloaded from http://heapro.oxfordjournals.org/ at Western Michigan University on March 25, 2016
