COM 408 Final Project

.docx

School

Arizona State University, Tempe *

*We aren’t endorsed by this school

Course

408

Subject

Communications

Date

Feb 20, 2024

Type

docx

Pages

Uploaded by cindyvsagastume on coursehero.com

1 COM 408 Final Project Cindy Sagastume Arizona State University

2 Introduction I was given the task to analyze a large dataset for The Tablet Company in order to find out who our target customers are that may be influenced into purchasing our new state-of-the-art tablet. Through analysis of the data provided, I will be able to highlight key information that may help the company better understand the population that inhabits the greater Phoenix-area and whether or not there is room for improvement in their marketing plans for this product. By the end of this report, I will ensure that my company knows exactly who to market to for the best and largest return on investment of a product in their product sales history. Descriptives The dataset collected consisted of 500 individuals. Out of the 500 individuals, 274 of them were women and 226 of them were male. The ages of these participants varied between 9 to 102 years old, with a median of 50 and a range of 93. The number of children these individuals reported having ranged between 0-12 which computed an average of 2.11 and a standard deviation ( SD ) of 1.89, as 132 of them have at least 2 children. With a mean of 51.66, 11.2% of these participants responded that they spend about 50% of their time out of their day using some sort of technology (phone, tablet, computer, gaming device). Through their preferred form of technology, 48.4% reported spending about 18 hours per week surfing the internet, along with a mean of 14.37. 48.5% participants reported spending about 4 hours per week just on emailing, which computed a mean of 6.23 and a SD of 9.88 . In addition, out of these 500 participants, 178 reported willingly watching at least 3 hours of TV per day. T-Tests In the first t-test that I conducted, I used “respondents sex” as my independent variable and “age of respondent” as my dependent variable. We found this test to be insignificant as we

3 found there to be similarity in the variances between the groups of ages of both males and females. With a mean difference of just 1.27, our female group came in higher with a mean of 52.85 and our male group just below them with a mean of 51.58. This means that both our female group and male group are roughly the same age, if not a year or two younger. There is a chance that this may have a negative impact on our companies sales goals as this age group may not be looking for new tech. In the second t-test that I conducted, I used “respondents sex” as my independent variable once again, and “number of children” as my dependent variable. This test was also considered statistically insignificant as there were, again, similarity in the variances between the groups of both males and females and the number of children they each reported. With a mean difference of .171, our female group came in higher with a mean of 2.19 and the males below them with a mean of 2.02. This information may benefit the company as they know that out of the 500 respondents, 26.4% of them have at least 2 children and can mean that they are likely to buy the tablet, not for themselves, but for their children (see figure 1). ANOVA In the first ANOVA test I ran, I used “did r vote in the last election” as my independent variable and “homosexuals should have a right to marry” as my dependent variable. Just by looking at the group’s means alone, I was able to determine that this test was found statistically insignificant before even looking at my value that measures significance. Within the three groups, people who were “ineligible” to vote came in at the highest group and had a mean of 3.51, people who “voted” had a mean of 3.50, and people who “did not vote” came in last with a mean of 3.39. In the second ANOVA test I ran, I used “political party affiliation” as my independent variable and “homosexuals should have a right to marry" as my dependent variable again. This test did prove to be statistically significant as the value computed through the

4 ANOVA test was below .05. When having compared all three groups through a Tukey test, I was able to see exactly where the difference was that was considered significant and found it to be between the Democratic party and the Republican party on whether or not they believe homosexuals should have a right to marry. The Democratic party had a higher mean of 3.39, while the Republican party had a mean of 2.63; the mean difference between the two being - 1.302. Bivariate Correlations In my first correlations test, the 2 variables I used were “email hours per week” compared to “www hours per week”. Out of our sample size of 497, I found this test to be statistically significant! My r value for this correlation was .281, which happens to be a positive correlation but weak in correlation strength (see figure 2). When computing the coefficient of determination, I got .078 or 7.8% which describes the overlap between the two variables. What this means is that as the hours of surfing the world wide web per week goes up, so do the email hours per week, or vice versa. In my second correlations test, the 2 variables used were “job satisfaction in general” compared to “how often r finds her work stressful”. This test was also statistically significant! Our r value for this correlation was -.198. This describes a negative correlation and slight correlation strength (see figure 3). The overlap between these two variables was .039 or 3.9%. This means that as the number for “how often r finds her work stressful” goes up, the values that measure “job satisfaction in general” goes down, or vice versa. Linear Regression In my first regression test, I used “age of respondent” as my predictor variable and “number of children” as my dependent variable. An effect size of 3.8% was computed as the overlap or variability between the two variables. This effect size was found to be statistically

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help