IAT&T; LTE 11:01 AM Recall that descriptive statistics help us better understand
ID: 3335613 • Letter: I
Question
IAT&T; LTE 11:01 AM Recall that descriptive statistics help us better understand the characteristics of our sample. Read the following passage below. Based on these frequencies, percentages, means, and standard deviations, form some conclusions about the sample. Are certain demographics under or oversampled? How might the descriptive statistics influence the ecological validity of the study's results? Explain your reasoning “178 individuals from Illinois started the online survey, but 40 participants exited the survey before completing half the items and were not included in the analysis, resulting in a sample of 138. The sample was primarily Caucasian (n=136, 98.6%) and female (n=107, 79.3%). Three-quarters of the participants had a bachelor's degree or higher (n= 103,75%). The mean age of the sample was 45.84 years (SD-13.76)," (Spialek & Houston, 2017, p. 12) ExplanationExplanation / Answer
The mean age of sample is 45.84 and standard deviation is 13.76. Since sd indicates the spread of the data around the mean we can say that the marjority(> 66%) of our 138 individuals are between the age 32 and 60 (45.84 +&- 13.76). This would mean that the older generation will have more say on the outcome of the survey rather than youngsters and teenagers. Unless it was the purpose of the survey to have opinion of older generation this would make the results more skewed towards them. Also 136 of the 138 participants were Caucasian. We can claim that they were oversampled. Normally we would like for all races to have an equal say on our result. However if about 98% of our sample consists of only one major race then the bias towards that race increases alarmingly which is not an ideal situation to face. One conclusion which could also come from this is that Illinois has very large no. of Caucasians which may not actually be true but would be implied by the sample. And while the number of females in the sample may seem large it is not that much of an obscure demographic to have. It would purely depend on the surveyor whether or not consider gender as a factor since we assume a sample to be random it is not impossible for one gender to have more influence than other. We also see that since three-quarters of the participants had bachelors's degree we will have more of an educated opinion in the survey but it may also overshadow opinions and problems of the less-educated.