Final Exam Study Guide: Part 1 - Technology

Your final exam is composed of two parts. The first part is to use Minitab to answer the questions. The second part will be more like what you have done on your other tests throughout the semester.

Both parts of the exam are open notebook. This may include your old tests, activities, technology projects, etc. You may not use your textbook. You may use minitab notes from the activities and tech projects on the first part of the exam. Since I asked you not to print these in the classroom to save paper, you may pull them up online and look at them during your test.

The following is not a question by question description of what you will need to know like most study guides. Part of this test is to see if you know where to go to do the things that are asked. Instead, this guide is designed to let you know which areas of Minitab you should be familiar with. I would certainly put this in your notebook to use as a reference during the exam.

File Menu

Open Worksheet. You will need to open a data set to complete the test.

You do not need to print or save your results. You do not need to copy and paste into Word. Write all answers on the exam.

Calc Menu

Calculator. You will need to be able to create a new variable based on an expression involving existing variables.
Probability Distributions.
- You will need to be able to work with the Binomial, Normal, Student's T, Chi-Square, and F distributions. You will not need to sample from columns.
- The "probability" option can be used to find probabilities of specific outcomes.
- The "cumulative probability" can be used to find the area to the left of a value. You'll have to manipulate the output if you need the area to the right. If you need a two-tail p-value then you'll have to manipulate the input and the output.
- The "inverse cumulative probability" will return the critical value that has the input area to the left. You'll have to manipulate the input if you are given the area to the right. If you need two critical values, you'll have to manipulate the input and run the procedure twice.

Stats / Basic Statistics Menu

Display Descriptive Statistics.
1 Sample T. Used to test a claim about a single mean.
2 Sample T. Used to test a claim about two independent means. The test mean is the difference in the means.
Paired T. Used to test a claim about the mean of the difference with dependent samples.
1 Proportion. Used to test a claim about a single population proportion. Remember to enter proportions as decimals. Also check the "base interval and test on normal distribution" box.
2 Proportions. Used to test a claim about two independent proportions. The test value is the difference in the proportions. Be sure to check the "use pooled estimate for p" box.
Correlation. Used to find correlation between two quantitative variables.
Normality Test. Used to check whether a sample comes from a population with a normal distribution.

Stats / Regression Menu

Regression. This will find the regression equation when there is significant linear correlation. Remember that when there is no linear correlation, the best estimate for a variable is its mean. You do not need to worry about residuals or fits for this test.

Stats / ANOVA Menu

One-Way. Used to test the claim that the three or more means are equal when the samples are independent. There should be one column containing the factor and one column containing the values to test.
One-Way Stacked. This is the same test, but each sample is in a different column.

Stats / Tables Menu

Cross Tabulation. This is used to create a contingency table and perform a test for independence. Be sure to check the chi-square box to do a test for independence.
- If you frequencies instead of each row being a separate case, then include the two classification variables and check the "Frequencies are in ..." box and put the frequency column there.
- If your data is such that each row is an individual case and there are no frequencies, then include the two classification variables and the variable you want to tally in the classification variables box. You will get a table for each value in the third variable listed. For example, if you have the three variables gender, party, and race and specify them in that order, it will give you two tables (one for whites and one for blacks) containing the tallys with gender for the row variable and party for the column variable.

Graph Menu

Plot. You need to know how to make a scatter plot. There will be three plots given on the test and you have to circle the right one. If you don't get one of the three, then you know you did something wrong.

Type of Questions to Expect

The questions are not like questions on previous tests where you have 13 parts and you have to go through and answer each part. Instead a sample question might be ...

Based on a p-value of _______, there ( is / is not ) enough evidence to ( reject / support ) the claim that less than 30% of the people smoke.

From this question, you need to know to go to the 1 proportion test, put in a claimed proportion of 0.30, change the alternative to less than, and then run the test. For questions like this, the p-value will be worth 3 points, the ( is / is not ) worth 1, and the ( reject / support ) worth 1. This p-value is the only thing you can't guess at and get right half the time, so that's how I'll know you've done the technology part of it.

Another question might read ...

Which of the following graphs is a scatter plot of the weight (y) vs height (x)?

This is a matching problem with three choices, so you could take a wild guess and hope you get it right or you could go to graph / plot and make the graph and compare it to one that's there. You could guess, but you lose five points if you guess wrong.

Notes

You will probably want a calculator.
You will definitely need Minitab.
Work through the problems you know how to do quickly. If you don't know how to do something, then skip it and come back to it later.
You will not be able to continue working on this exam in the testing center, they don't have Minitab loaded. Use your time wisely.
The exam is worth 100 points, each individual part on the exam is worth 5 points.
Half of the questions on the test involve finding a p-value or a confidence interval and making a conclusion.