|
The title page of the questionaire welcomes the responde nt to the survey and explains what the collected information will be used for. The researcher is introduced and they are informed that their answers will be kept strictly confidential. There are four sections to the questionaire. The first section measures change in brand preference on a pre-to-pos ad exposure constant sum scale. First, the respondents are asked to rate their brand preference for each brand by the liklihood of purchasing the brand. They must use whole numbers on a scale of 1 to 10. The same questions are asked about bottled water and juice. Next, the respondents are shown an advertisement of each brand of soft drink. They are asked to briefly view each ad. They then are asked to, once again, indicate which brand they are more likely to purchase. The second section measures the atttude towards each brand. 10 Likert items are presented for each of the three brands. The respondents are asked to indicate how strongly they agree or disagree with each statement about the brand. A value of 5 is given to "strongly agree" and a value of 1 is given to "strongly disagree." The third section is a set of 10 questions about the ad itself and the ad's message. These are 'yes' or 'no' questions. 'Yes' answers are given a value of 1 and 'no' answers are given a value of 0. These values are used for statistical purposes. This measures the attitude toward the advertisements. The last section is used to collect demographic information. These questions include sex, age, how many soft drinks they drink each day, why they drink soda, and where they drink it. This information is used to see if there is a correlation between the demographics and how each respondent reacts to the ad.
This survey follows a psuedo-experiment format, meaning there is no control group. Instead, the comparison is made between pre-to-post advertisement exposure. By measuring the difference between the two, the constant sum scale is used to measure the net effects of advertsing.
Due to time and monetary constraints, a non-random study was conducted. A sample size of atleast 60 was determined to be enough to preform statistical analysis. This sample size allows enough representation so that if statistical significance is found, projection can be made in 85 or more samples of 100 samples within the same population as these 64 were drawn, we would find results of the same magnitude. The survey was distributed by email to personal friends and acquaintances of the researcher. The email prompted the reader to visit the survey via the provided url, and asked to send the survey along to others who may be interested in participating. All respondents agreed to participate in the study by clicking the submit button at the end of the questionaire. A total of 78 respondents took the survey. However, only 64 of those questionaires were fully completed. Thus, the sample size of this research is 64.
The data for this survey was collected electronically from September 21, 2005 through October 5, 2005. Each email contained an url link to the questionaire. Once the respondent enters the questionaire, and answers all the questions, they click the submit button to complete the survey. The information is stored in a database file in the Center for Interactive Advertising. Both the database file and the online questionaire were connected through a cold fusion file, which contained a page that thanked each respondent for their participation in the survey. 64 surveys were completed in full, thus those are the ones that are used in this research. |