Statistical Analysis

Basic Statistics  

The purpose of statistically analyzing this particular data set is to determine how three divergent print advertisements, within the chewing gum product category, affect consumer preference, perception, and likeability toward three discrete chewing gum brands: Winterfresh, Big Red, and Orbit. In order to successfully collect and gather data, an online survey was distributed via e-mail, which garnered a total of 83 respondents. Here you will find the results from several statistical analyses performed on this data.

Part One. Correlated t-Tests

Brand Index Scores			Paired t-Tests
Brand Name	Mean Brand Index Score	Standard Deviation	Paired Samples	t Ratio
Winterfresh	33.8	4.7	Winterfresh — Big Red	2.59*
Big Red	31.9	5.5	Winterfresh— Orbit	2.57*
Orbit	36.2	6.2	Big Red — Orbit	4.11*
sample size (n)=83			* indicates p ≤ 0.15

Statistical Interpretation

According to the graph above, it appears that we can project the results calculated to the entire population because the three t-ratios are all significant. The t-ratios are said to be “statistically significant” because it is very rare that we would have collected these numbers by mere chance. These results imply that in 85 or more samples out of every 100 samples drawn from the same population as this sample of 83 individuals, we would expect to find the difference between the Brand Index Score means for each pairing of chewing gums (Winterfresh – Big Red, Winterfresh – Orbit, and Big Red – Orbit) in the population respectively.

Advertising Implication

This data suggests that the 83 respondents have a more favorable perception of and brand preference for Orbit than Winterfresh or Big Red. Big Red is the least liked chewing gum with a mean Brand Index Score of 31.9, while Winterfresh falls in the middle of the two. Though there is a slight difference between the Brand Index Score means, all three are relatively close, indicating that there is not a great difference in preference, though many favor one brand over the others.

Part Two. Between Groups t-Tests

Winterfresh Brand Index Score: Between-Groups t-Test
Pre-Post Change Score	Sample Size	Mean Score	Standard Deviation	t Ratio
Moved Up	28	34.6	3.2	1.11*
Moved Down	20	33.3	4.4
Winterfresh Ad Index Score : Between-Groups t-Test
Pre-Post Change Score	Sample Size	Mean Score	Standard Deviation	t Ratio
Moved Up	28	5.5	3.1	2.00*
Moved Down	20	3.6	3.3

* indicates p ≤ 0.15

Statistical Interpretation

Since Winterfresh’s Advertising Index Score and Brand Index Score are both significant at the α = .15, we can conclude that in 85 or more samples out of every 100 samples drawn from the same population as this sample, the Advertising Index Score population mean for the up-movers would be 5.5 and the Brand Index Score population mean for the up-movers would be approximately 34.6; the same as the AIS and BIS sample means found in the table above. Since the data hold a high level of significance, these means could be projected onto the population as a whole.

There is a significant difference in the mean BIS and AIS scores among the people who went up (viewed Winterfresh more positively after seeing the brand’s ad) and the people who went down (thought less of Winterfresh chewing gum after viewing the brand’s advertisement). The AIS score was more significant than the BIS score because the calculated t-ratio (= 2.00) has a smaller probability of occurring by chance.

Advertising Implication

According to these numbers, the up-movers had a more favorable opinion of the Winterfresh brand and advertisement than the down-movers. These results can be projected to the population and also illustrate that more respondents moved up than down after viewing the Winterfresh advertisement, although the two differ by a minute amount. It can be said that the print ad was liked by a majority of the respondents, and that the advertisement was more of a success than a failure.

Part Three. Chi-Squared Significance Test

Pre to Post Change Score		Brand Index Score Above Median (34.0)	Brand Index Score Below Median (34.0)
Moved Up	Count	16	12
	% Row	57.1%	42.9%
	% Column	43.2%	26.1%
	% Total	19.3%	14.5%
Same	Count	15	20
	% Row	42.9%	57.1%
	% Column	40.5%	43.5%
	% Total	18.1%	24.1%
Moved Down	Count	6	14
	% Row	30.0%	70.0%
	% Column	16.2%	30.4%
	% Total	7.2%	16.9%

 

Chi-Squared

3.55*

* indicates p ≤ 0.15  

Statistical Interpretation

The chi-squared significance test illustrates that there is a significant relationship between up, same, and down movers and being above or below the median Brand Index Score for Winterfresh. The smaller the chi-squared value is the stronger a relationship is between the up, same, or down-movers and being either above or below the median. The Brand Index Score median is used to show an equal number of people above and below the midpoint.

In 85 or more samples out of every 100 samples drawn from the same population as this sample of 83 people, we would expect to find about the same percentage distribution in the entire population as we have found in this sample. Since the data holds a high level of significance, these cell proportions can be projected onto the population considered.

Advertising Implication

The percentages calculated were not surprising. From the sample of 83 respondents, more respondents with above-median brand perceptions (43.2%) had improved brand perceptions after ad exposure than respondents with below-median brand perceptions (26.1%). 40.5% of the individuals questioned stayed the same, which is a rather typical outcome. 1/3 of the individuals questioned did not move either up or down, showing that there perceptions were not altered negatively or positively after seeing the advertisement for Winterfresh. It is interesting to point out that of the people who stayed the “same”, 57.1% were below the Brand Index Score median, illustrating that this group had a mediocre opinion of the Winterfresh brand before viewing the print ad, yet there perception of the brand stayed the same after looking at the advertisement.

Part Four. Frequency  

Pre to Post Change Score	Winterfresh		Big Red		Orbit
	Frequency	Percent (%)	Frequency	Percent (%)	Frequency	Percent (%)
Moved Up	28	33.7	16	19.3	15	18.1
Same	35	42.2	48	57.8	42	50.6
Moved Down	20	24.1	19	22.9	26	31.3

sample size (n) =83

According to the data, it is apparent that Winterfresh had the most individuals favor the brand more (33.7%) after viewing the magazine advertisement. Big Red had the largest amount of respondent attitudes stay the same (57.8%), which is not necessarily a negative outcome, but does illustrate that there is room for creative improvement. Orbit had the largest amount of individuals move down pre to post advertisement screening. This could be explained by respondents having a very high perception of the Orbit brand before viewing the print ad, that after examining all three ads, some respondents wanted to distribute a few points amongst other brands (possibly Winterfresh).

Part Five. Frequency Count

Brand Comparison	Frequency	Percent (%)
Winterfresh > Big Red	34	41.0
Winterfresh ≤ Big Red	49	59.0

sample size (n) =83

The Frequency Table shows that 34 people have a higher opinion of the Winterfresh brand than the Big Red brand, with these 34 people representing 41% of the sample. This output states that out of 83 total respondents, 41.0% of the sample rated Winterfresh higher than Big Red, believing that Winterfresh is a better brand compared to the competitor. The data also shows that 49 individuals (59% of the sample) believe that Winterfresh is either a lower or an equivalent brand to Big Red.

Part Six. Simple Correlation Coefficient

Comparison of Brand Index Scores	Sample Size	Correlation (r)
Winterfresh and Big Red	83	.05*

* indicates p ≤ 0.15

Statistical Interpretation

The correlation between the two brands Brand Index scores is .05, indicating that Winterfresh and Big Red have a very low positive correlation. The low positive correlation illustrates that if one brand index score goes up, the other brand index score also increase, but because there is such a little positive correlation, there is too small of a relationship between the two to discern. Because the correlation coefficient is statistically significant, we are able to confidently project the findings onto the population as a whole. In 85 or more samples out of every 100 samples drawn from the same population as this sample of 83 respondents, the correlation of the two brand index scores for Winterfresh and Big Red would be approximately the same as found in this sample.

Advertising Implication

This correlation coefficient tells us that there is really no relationship between how people view the Winterfresh print ad and how they view the Big Red advertisement. If a respondent examines one ad and then looks at the other, they do not judge the later based on their opinion of the first ad observed.

Part Seven. Female Sub-Sample Frequency  

Advertisement Characteristic	Winterfresh		Big Red		Orbit
	Frequency	Percent (%)	Frequency	Percent (%)	Frequency	Percent (%)
Lame	16	32.7	19	38.8	14	28.6
Not Lame	33	67.3	30	61.2	35	71.4

 

female sample size (n) =49

 

In the table above, the respondents questioned were female, and were asked if they considered any of the three different either “lame” or “not lame.” It is shown here that out of all three brands, Big Red’s advertisement (38.8% of the sample) was considered more lame than the other two rival brands. Orbit’s print ad was believed to be the least lame out of all three brands; with 71.4% deeming the ad “not lame”, while 28.6% thought it was “lame.”

Though I did not find these results surprising, I did find this question and data output interesting. It is clear that many females disliked the new Big Red gum pack ad because the gum parodied a pack of cigarettes. Many might find this distasteful in an age where cigarette smoke and its harmful effects are frowned upon. Though this ad may offend some, I believe that those who are insulted by the print ad are the individuals that Big Red does not covet as customers.