Reference no: EM133163334

Instruction

The Happiness and Engagement Dataset

Assignment Content

Now that you've run descriptive statistics with your data, it's time to create a hypothesis and test your hypothesis. This part of the Statistics Project will take you through the process of creating and testing your hypothesis through statistical methods, using Microsoft Excel. Creating hypotheses provides you the opportunity to think like a researcher and help you understand and critique research articles you read.

For Week 3, your hypothesis and statistics should be appropriate for testing means of two groups.

State a hypothesis for the Happiness and Engagement Dataset from Part 1 of the Statistics Project. Your hypothesis can be anything based on the variables you have in your dataset. One example: Teaching Method X provides higher test scores than Teaching Method Y.

State the null hypothesis. (Example: Teaching Method X scores are equal to Teaching Method Y scores.)

Identify and justify which type of statistical analysis will be appropriate for this data.

Review the steps beginning on p. 202 in Statistics Plain and Simple describing how to run an independent samples t test.

Run an independent samples t test on the data in your dataset in Excel.

Write a 125- to 175-word summary of your interpretation of the results of the t test, and copy and paste your Microsoft Excel output below the summary.

Format your summary according to APA format.

If we try to use t tests and ANOVA by just comparing the means of two variables, that doesn't tell us much, other than whether the numbers of scores in each group are different from the other scores in the other group. So, you CANNOT just compare the scores for Supervisor with Telecommute and think this is doing a t test or ANOVA! So don't make this mistake!

With either test, ANOVA or t tests, the idea is that we are comparing the means of two groups with t tests and more than two groups with ANOVA, on a particular dependent variable - which is what we want to measure.

So, we start with grouping variables (which will be nominal scale data) like gender, telecommute, relationships with coworkers, etc. as your factors or independent variables. These are our factors, or "treatments." In this case we are not actually doing anything to the groups, but rather we are taking naturally occurring groups. Then, we want to see if the mean scores are different on a particular continuous variable (which means it will at interval or ratio level data) such as rating scales for happiness or engagement - this is the dependent variable, what we are measuring and trying to find differences on these scores between our different groups.

So, for your t test, if your independent variable is gender, and your dependent variable was"happiness", use the sort function, and create two new variables, onewith the happiness scores for all males and one for the happiness scores forall females. Then run the t test comparing your new variables. This enables youto compare the means of happiness (the dependent variable) of your two groups(in this case, male and female).

• NOTE on Sorting in Excel and Running a t test: When you sort one variable (one column) in Excel, it not only sorts that column, but all the other variable columns as well.

• So you would highlight one of your grouping variables like Coworkers, then in "Data" run the A to Z sort, which gives three sets of scores (1, 2, and 3s, all arranged in the new single column. Sometimes it works better to do the A to Z sort first, then select your variable from a pop up window. NOW, when you sort by your categorical independent variable, the other columns of variables will correspond, creating a newly sorted columns for a dependent variable that is arranged according to the Coworker (or other variable) scores. So, you then go to your dependent variable column (like happiness) and highlight the scores that are across from each group of numbers from the coworker column. It helps to move the two columns together so you can easily see what you are doing.

• Then, from you Happiness column, you will copy and paste into two new columns. These are the two new groups (levels) that you will compare on Happiness.

• Choose the test: t tests. Usually, you will use t test assuming equal variances. Do not use Pared samples unless you have sound reasons for this choice. In the "Input Range" just highlight all your columns, and make sure "Columns" is checked. Click OK, and you will obtain an t score as the test statistic, and then again, a p value that will tell us the significance level.

Attachment:- Hypothesis Testing.rar