Analyze 3 answers, just like before!

1)

BERNIE

Part One - Hypothesis Testing

We use hypothesis testing to validate an assumption and testing the hypothesis helps one make an informed decision. The five steps for hypothesis testing are:

1. Making assumptions
2. Stating the research and null hypotheses and selecting (setting) alpha
3. Selecting the sampling distribution and specifying the test statistic
4. Computing the test statistic
5. Making a decision and interpreting the results

By using these five basic steps, it will help you greatly in hypothesis testing. It gives you a procedure to follow, regardless of the particular problem you are working with. I believe the trickiest part of the hypothesis testing are the outcomes.The null hypothesis is the statement being tested. The alternative hypothesis is the statement you want to be able to conclude is true. One example of the hypothesis testing that should be appropriate would be the cost of living. Where is it cheaper to live Seattle or Portland?That is something that I recently just figure out when I was looking to move.At the end, I concluded that neither was cheaper and we moved somewhere else.

Part Two - T-tests

From the T-tests we can conclude that the salaries of the male and female employees are not equal. As there is a good bit of information to use, but we don't have their individual job descriptions.

Part Three - F-test

It makes it more complicated because there is that gray area of every scenario put into play. All the scenarios that are a part of the gray area of question get put into the mix and it makes the final results look questionable.All the data that is being used needs to have correct information that is current and truthful along with the right sources. If the information is not fully correct then that leaves all of the variations null and the test was a waste of time.Some things that could impact variations on salary could be the position that you are in. It all depends also on the department you are in as well. With the Government, there are different pay structures. So even though you do the same work as others sometime they get paid more than you. It also depends on what department you are in as well.

2)

JUAN

Part One - Hypothesis Testing

The hypothesis testing procedure is designed to ensure that data is analyzed in a consistent and recognized fashion so everyone can accept the outcome.

What is Five-step procedure for hypothesis testing:

• Step 1: State the null and alternate hypothesis
• Step 2: Form the decision rule
• Step 3: Select the appropriate statistical test
• Step 4: Perform the analysis
• Step 5: Make the decision, and translate the outcome into an answer to the initial research question.

What does it do for us?

We can conduct a test accurately so that we can answer specific questions:

1. If some comparison is statistically equal or not.
2. To make sure the difference is not a sampling error.
3. To make sure the difference is statistically different, which is not caused by chance.
4. Most importantly, to test if two things are related to each other. For example, is the salary of male statistically different than the salary of female, if the answer is yes, we can conclude that gender is a significant factor that influences the salary level.

Why do we need to follow these steps....

These steps allow us to answer questions above accurately. It allows for there to be a structure when making an informed hypothesis

What are the tricky parts? (Rules!)

We need to use a directional set of hypothesis statements to answer a question about the directional difference.

The null and alternate hypothesis statement must, between them, account for all possible actual comparisons outcomes. (Page 3)

The variables must be listed in the same order in both claims.

The null hypothesis must always contain the equal (=) sign.

The null can contain an equal (=), equal to or less than (<=) or equal to or greater than (=>) claim.

.

This could be presented on a hypothesis based to on the probability of anyone having a car accident based on the key factors such as income and career. Careers with high-income vs. Careers with lower income.Income could determine the type of car they drive (old vs. new or economy vs. luxury).

Part Two - T-tests

Since we rejected the null hypothesis in both approaches (and both will always provide the same outcome), we can answer our question with No - the male and female mean salaries are not equal. We only conclude that the salary difference between man and woman is statistically different, meaning this difference is not caused by chance. But, we did not use any statistical tools to prove that male and female are performing the equal work. We would prefer use T-test to prove if the difference of economic value added per hour between male and female is statistically significant. If no, we can conclude man and female do not receive equal pay for equal work.

The results are missing information regarding whether the data has been controlled for external factors which may be statistically significant. If for example, the data for male salaries was taken from geographic areas with a high cost of living which could consequentially have on-average higher salaries, and the data for female salaries was taken from the opposite areas, then the conclusion drawn here could be invalid. Additional tests would need to be performed that analyze other demographic factors, and the interaction between these demographic factors.

Part Three - F-test

Why does variation make our analysis of the equal pay for equal work question more complicated?

Statistical data and tests are based on probabilities; there is a possibility that the wrong decision could be made. If the variation in one set of salary data is significantly greater than the other, then concluding using a t-test may not be valid. This would likely indicate that the researchers need to obtain more salaries to reduce the variation or control for external factors which can be causing the variation.

What causes of variation impact salary that we have not discussed yet?

Other factors that can play a big part in determining how much someone is paid can be described as seniority and education. These variations have not been added to the calculation and have not yet been discussed. Someone with a masters degree could potentially be getting paid with someone with no degree as well as someone with a bachelor's

.

How can you relate this issue to measures used in your personal or professional lives?

The difference between genders is an interesting factor in my line of work. My job is composed of mostly men in the customer service environment. We tend to debate who has more issues that require customer service assistance - men or women. To analyze this, we could compare some customer service requests in a day against gender. However, if there were too much variation in this data, we would need to analyze other influencing factors, such as the customer's economic status.

3)

SHERRY

Part One - Hypothesis Testing

What is this?

The hypothesis testing procedure is designed to ensure that data is analyzed in a consistent and recognized fashion so everyone can accept the outcome. It makes assumptions about a how things work from data provided.Deciding which testing to use depends on the question being asked.

The procedure itself has five steps:

Step 1: State the null and alternate hypothesis

The null hypothesis is the "testable" claim about the relationship between the variables.We willreject the null hypothesis when the p-value is

equal to or less than 0.05 (this probability is called alpha). Other common values are .1, and .01

Step 2: Form the decision rule

This step involves selecting the decision rule for rejecting the null hypothesis claim. We will always test the null hypothesis before deciding the level of evidence.

Step 3: Select the appropriate statistical test

Selecting the appropriate statistical test is the next step.

Step 4: Perform the analysis

Performing the analysis comes next.All the math part involved is in Excel so there is no calculations by hand.

Step 5: Make the decision, and translate the outcome into an answer to the initial research question. Interpret the test results, making a decision on rejecting or not rejecting the null hypothesis, and using this outcome to answer the research question is the final step. (Tanner, Youssef-Morgan, 2013).

What does it do for us?

Hypothesis testing allows us to look at a sample rather than the entire scenario.It allows you focus on graphs and concepts rather than numerical data.

Why do we need to follow these steps in making a judgement about the populations our samples came from?

I think it to make us look at the sample in a different way.It allows you to decipher each piece of the puzzle.Hypothesis testing lets you ask the question, "is it large enough to be a significant factor?"

What are the "tricky" parts of developing appropriate hypotheses to test?

I would say that the tricky parts of hypothesis testing are the outcomes.If the entire sample is not tested then there will not be an accurate outcome.

What examples can you suggest where this process might be appropriate in your personal or professional lives?

In my personal life, I would probably use hypothesis testing at my church with the youth group. I am interested to know exactly how many participate in different activities and at what times they do.I do not need the whole number of youth at this time just a certain age range.

Part Two - T-tests

ReadLecture Five. Lecture Five illustrates several t-tests on the data set. What conclusions can you draw from these tests about our research question on equal pay for equal work? What is missing from these results to give us an answer to the question? Why? (This should be started on Day 3.)

According to the t-tests, it showed that the male salaries are not the same as the female salaries because of population change. The test is not analyzing the entire company but only a small portion.There are several things missing from these results.Are the males and females doing the same job? Does one sex have more seniority than the other?Once questions such as these are answered then we can get a better understanding of the results.

Part Three - F-test

ReadLecture Six. Lecture Six introduces you to the F-test for variance equality. Last week, we discussed how adding a variation measure to reports of means was a smart thing to do. Why does variation make our analysis of the equal pay for equal work question more complicated? What causes of variation impact salary that we have not discussed yet? How can you relate this issue to measures used in your personal or professional lives? (This should be completed by Day 5.)

Variation is prevalent in all of our lives.I have been using variation at home with our water bill, but didn't realize it.I am still trying to figure out how the dollar amount is calculated.There was two months when the usage was exactly the same but one month the bill was higher than the previous one.Variation makes it more complicated because there are so many unanswered questions involved.

When performing statistical tests we are basing it on probabilities.Mistakes can be made which would throw everything out of order.One of the main things that I think has not been added to the calculation is seniority and the different jobs between male and female.