With the basic understanding of the survey project, I think it's ready to describe what I would like you to do

when solving P2. Please attend to the "2024_FinalEam_AnalyzingSurveyData.sav" file when working on P2.

We need to first "clean" the data set by conducting a few straightforward tasks before subsequent analysis.

Page3

1. Step #1: Please focus on Q5_5. Specifically, responses with values of 3, 4, or 5 on Q5_5 should not

be included in the subsequent analysis to avoid biased outcomes. Why? During the clarification on

how to answer the online survey, I mentioned that the "correct" response to this attention check

question should be either 1 (strongly disagree) or 2 (disagree). Therefore, those who selected "3,"

"4," or "5" did not pass the attention check and should not be included in the subsequent analysis.

In addition, when we create a composite variable to represent respondents' perception toward

video lecture, we should skip Q5_5 (see a snapshot shown on page #2) and focus on Q5_1, Q5_2,

Q5_3, Q5_4, and Q5_6). See Step #2 for more details.

2. Step #2: Let's attend to the following five questions: Q5_1 ("The video lectures in the Quantitative

Analysis for Business course ...were helpful in learning the data analysis materials"), Q5_2, Q5_3,

Q5_4, and Q5_6 ("The video lectures in the Quantitative Analysis for Business course ... I would

recommend video lectures for other similar online classes"). Specifically, we need to generate a

new variable called "video," which will represent the mean value of Q5_1, Q5_2, Q5_3, Q5_4, and

Q5_6.

3. Step #3: The question item Q11 is a binary variable with two values (1 for male and 2 for female).

Disregard the responses with values other than 1 or 2 in Q11 (in other words, focus on the cases

whose Q11 score is either 1 or 2).

4. Now, please employ a ttest to examine whether the mean score of the newly created variable

named "video" is statistically different between male and female respondents.

a. Please analyze whether the mean score of your newly created variable, "video," differs between

genders. Specifically, please calculate and report the mean score of "video" for both females and males.

Subsequently, conduct a suitable ttest to determine if there is a statistically significant difference between

the mean scores of "video" for the two genders. Finally, interpret the findings in straightforward language

b. Does the newly created variable, "video," show any outlier(s) by verifying if the values of the

standardized variable are beyond +/3.29? Please identify the value(s) of the "video" variable associated

with the outlier(s) and their corresponding ID number(s). In addition, please also identify the maximum and

minimum values of the standardized variable along with their corresponding "ID" numbers. In case you find

no outlier, feel free to put NA (not available) in the following table