+1-415-670-9189
info@expertsmind.com
Discuss issue of statistical power in non-parametric tests
Course:- Basic Statistics
Reference No.:- EM13779034




Assignment Help
Expertsmind Rated 4.9 / 5 based on 47215 reviews.
Review Site
Assignment Help >> Basic Statistics

Question: You will submit one Word document for this assignment, In the first part, you will provide short answers to the following questions.

Begin by downloading the following SPSS datasets:

. Activity 6a.sav (found on the "additional resources" page)

. Activity 6b.sav (file you created in Week 8)

. Activity 6c.sav (found on the "additional resources" page)

Part A. Questions about non-parametric procedures

What are the most common reasons you would select a non-parametric test over the parametric alternative?

Discuss the issue of statistical power in non-parametric tests (as compared to their parametric counterparts). Which type tends to be more powerful? Why?

For each of the following parametric tests, identify the appropriate non-parametric counterpart:

Dependent t test

Independent samples t test

Repeated measures ANOVA (one-variable)

One-way ANOVA (independent)

Pearson Correlation

Part B. SPSS Assignment

In this part of the assignment, you will perform the non-parametric version of the tests you used previously. In each case, assume that you opted to use the non-parametric equivalent rather than the parametric test. Using the data files from earlier activities, complete the following tests and paste your results into the assignment Word document:

Activity 6A: non-parametric version of the dependent t test

Activity 6B: non-parametric version of the independent t test

Activity 6C: non-parametric version of the single factor ANOVA

Answered:-

Verified Expert


Preview Container content

Part A

In order to determine whether the researcher claim is significant or not, he need to perform a statistical test. Choosing an appropriate statistical test is necessary to support the researcher’s claim. The statistical tests can be classified into two categories, namely,
Parametric tests, and

Non Parametric tests

Parametric tests are tests which deal with parameters of the population like, mean, standard deviation, etc. If the parameters of the population are known, then, we can perform the parametric tests; else, we need to proceed with non parametric tests. In order to perform the parametric tests, we should validate the sample data using some specific tests, called, homogeneity, normality tests and independence. If all these assumptions are satisfied, then we can say that the distribution of the parent population follows normal and therefore, the sample data is highly reliable to perform parametric tests like t test, Z tests, F tests etc.

If these assumptions are violated, then it is inappropriate to perform the parametric tests. In these kind of situations, we need to perform non parametric tests. The non parametric tests are also called as distribution free test as it do not depend upon any parameters of the population. Non parametric tests are applicable to both variables and attributes, while parametric tests are applicable only to variables. For variables with nominal and ordinal data, there are no parametric tests available, but non parametric tests can be used for these kinds of data too.




Put your comment
 
Minimize


Ask Question & Get Answers from Experts
Browse some more (Basic Statistics) Materials
The manufacturer of Winston Tire Company claims its new tires last for an average 40,000 miles. An independent testing agency road-tested 20 tires to substantiate the claim
Convert the frequency distribution to a cumulative frequency distribution. Develop a cumulative frequency polygon. About 60 percent of the orders were filled in less than ho
Is the distribution of scores at this high school significantly different from the national results?- Was there a significant different between the performances of boys and gi
What is the probability that a randomly selected household with a head of household in this age group had less than $70 000 in life insurance coverage?"
A coin was flipped 100 times and came up heads 59 times and tails 41 times. The expected outcome was to land on heads or tails 50 times each.
A real estate company wants to study the relationship between house sales prices and some important predictors of sales prices. Based on data from recently sold homes in the
In a random sample of 120 cheerleaders, 54 had suffered moderate to severe damage to their voices. With 90% confidence, what can we say about the maximum error if we use the
Random Sample Obtain (generate) a random sample of 180 numbers between 1 and 60. You may use a calculator, excel, or a table of random numbers to achieve this. Summarize you