**StudentID: **101736154

**Nickname: **mercury

**Q3: **D

**Honorcode: **1

**Date: **3/15/00

**Time: **12:55:27 PM

For all four test it is assumed that the population from which is the sample is drawn is normally distributes, and therefore, the formula for all four of these tests are quite similar. In all four tests, the formula is the difference between two means, whether it be between 2 samples, a sample and the population, or 2 populations; divided by the standard deviation for the sample.

If you were using Pearson's r to evaluate the data, in the description of the design you would look for something that refers to finding the relationship between two variables (A & B), in one sample of data. If you were to use and independent t test to evaluate the data, in the description of the design you would look to see that A and B were two separate groups in an experiment, that have separate independent variables, and have been chosen separately from each other, with one dependent variable being measured at the end of the study. If you were to use a dependent t test to evaluate the data, in the description of the design you would look to see that each observation in one sample is paired on a one to one basis through either matched measures, repeated measures design, or pre-post design, with a single observation in the other sample.

**StudentID: **101780234

**Nickname: **lucky

**Q3: **D

**Honorcode: **1

**Date: **3/15/00

**Time: **1:32:35 PM

All of the formulas include the difference between two means in the numerator,[ (m-M), (x-M), (md-Md), and (M1-M2) respectively], and the standard error in the denominator,[ (SD/(n^1/2)), {(((x-M)^2)/(n-1))^1/2)/(n^1/2)}, {(((D-Md)^2)/(n-1))^1/2)/(n^1/2)}, and {((Sp^2/N1)+(Sp^2/N2))^1/2} repectively ].

1) Use Pearson's r if there's one group in the test and there's a relationship between the 2 variables, 2) Use an independent t test if there are two groups in the study and the two groups are independent of each other, 3) Use a dependent test if there are two groups in the study and the two groups are related.

**StudentID: **101186427

**Nickname: **Spunk

**Q3: **D

**Honorcode: **1

**Date: **3/15/00

**Time: **1:42:23 PM

The formula for finding both the variance and the SD in all of these tests is somewhat alike. To find the variance for a z-test you use SD^2=E(x-M)^2/N. This formula tells you to take the sum of squared deviations from the mean and divide it by the number of subjects. To find the SD all you need to do is take the square root of the variance. The formula for variance in a one sample t-test is only slightly different. S^2= E(x-M)^2/N-1. The only difference is that you divide by N-1 instead of just N. When using a t-test you have to have the degrees of freedom to find the cutoff score for a hypothesis test. In this case, the df= N-1. The estimated population SD is the square root of the estimated population variance, which is similiar to the process for finding the SD in a z-test. The t-test for dependent means is exactly the same as the t-test for a single sample except that you use something called difference scores and you assume that the population mean is zero. When doing a dependent t-test you are given two scores from the same subjects, like a before and after score. Before you can calculate the variance and standard deviation you need to use difference score, which means you subtract one score from the other for each person. The formula is the same except you are now usually the differences scores. S^2= E(D-Md)/N-1. THe D is X1-X2. Basically you would now do the same process as before but this time using the differences scores. The df= N-1 which is the same as the one sample t-test. To find the Standard Deviation you would take the score root of the variance. Lastly there is the independent t test which gives you two completely independent groups of people. The main difference between this and the other t-tests is that this ones key result of the study is a differene between the means of the two samples. When calcualting the variance and Standard Deviation you need to first calcualte the variance using the one sample t-test formuala for each separate group. You will get two different variance. THen you need to find the S^2p which is the pooled variance which is the weighted average of sample variances. This is where the process gets different. Another formula that is similiar in each test is the formula for calcualting both the z-test and the t-test. The formula for a z-test is Z=X-M/SD, the formula for a onesample t-test is T=M-U/Sm, the formual for the dependent t-test is t=(Md-0)/Sm and finally for the independent t-test is t=M1-M2/Sdif. Basically you are subtracting two means from each other and then dividing by the standard deviation. However, for a one sample and a dependent t-test you need to divide by the standard error. Also, for the independent t-test you are dividing by the standard deviation of the differnece.

You know to use a Pearson test when the problem is asking about a relationship between two variables. In other words, the problem will ask about the degree of correlation, which is the extent to which there is a clear pattern of some particular relationship between two variables. For a one sample t-test, you will be comparing the mean of a single sample to a population with a known mean but unknown variance. You use a dependent t-test when you know neither the population mean nor its variance. It will also describe a repeated measure design, where each person is measured more than once. When using a independent t- test you will be comparing two completely independent groups, a control group and a experimental group. You will be comparing the means of these two entirely separate groups of people.

**StudentID: **101566449

**Nickname: **Peaches

**Q3: **D

**Honorcode: **1

**Date: **3/15/00

**Time: **4:14:41 PM

They all rely on random samples that are normally distributed. They all in some way use the number of subjects to get the number in the denomanator. Both the z-test and the t-test use only one sample. All but the z-test use an estimate of sigma because it is not given. Both the Dependant and the Independant t-tests rely on 2 random samples. In all of the tests you are subtracting the mean or u and dividing by some sore of standard devieation(estimated or not).

If the description of the design implied a relationship between the two variables that may be corralated, I would use Pearson's r. IF the description said that the two variables were not related and it wanted to measure the dependent variable at the end of the study I would use an independent t. If the description did however tell me that the two populations were in some way connected and we wanted to see of the dependant variable affects an independent variable with in one of the population I would then use a dependent t test to evaluate the data.

**StudentID: **101115311

**Nickname: **scoobs

**Q3: **D

**Honorcode: **1

**Date: **3/15/00

**Time: **9:39:46 PM

The formula for the z-test involves taking the sample mean minus the mean of sample means/population mean, divided by the standard deviation of means. The one sample t-test is similar to this because it also begins by taking the mean of the sample minus the mean of the population. It differs because since the st. dev. is unknown, it divides by an estimate of the st. dev.(s/n^1/2). The dependent t-test takes the mean of the sample of the difference scores(MD) minus the mean of the population of the differnce scores(equal to 0). This number is again divided by an estimate of the st. dev. This estimate however is computed through using the st. dev. of differences. In the independent t-test we deal with the difference between the two sample means again, divided by the st. dev. o f differences. Similarities in all four tests include that we use the sample means(although some are means of differences) as well as the st. dev. or an estimate of the st. dev.

1)You would know to use a pearson's r test if the description of the design talked about a relationship or correlation between the groups. 2)You would know to use an independent t-test if we want to compare the means and we see that the groups are chosen independently of each other. The groups are made up by two entirely separate groups of people whose scores are independent of each other. 3)We would know to instead compare the samples by using a dependent t-test if the groups were related. For example if they were paired/matched samples, if they came from just one group of individuals in a pre-post test situation, or if they were from a single group of individuals who took the test twice in a repeated measure situation.

**StudentID: **101133209

**Nickname: **Elon

**Q3: **D

**Honorcode: **1

**Date: **3/15/00

**Time: **10:38:46 PM

All four of these tests are inferential statistical tests that use at least one sample. Each test can be used in a situation to help the psychologist perform a hypothesis test. With using the appropriate test at the appropriate time, the psychologist can make a decison regarding whether or not the results are statistically significant. It is that decision which helps the psychologist make conclusions regarding the experiment in general.

In the description of the design, I would look to see what question the experiment was trying to answer. I would use Pearson's r if the experiment was being run to determine if there was a relationship or not, or how significant that relationship was. Another way to know to use Pearson's r is when the experiment is not asking if there is a cause and effect relationship, since this test cannot determine that information. If I was looking to use the independent t test I would look for a situaiton where two distribution of means were being evaluated. An example would be the mean loss of weight in each population or the mean stress score before and after a vacation. In each of these situations the groups are independent of each other. Finally, if I wanted to use a dependent t test I would look for the two groups to be related in some way, such as being matched with each other, or a test being repeated on the subjects, or a test being given before and after a given situation. In this case there is really one observation being made regarding the differences between each group.

**StudentID: **101385622

**Nickname: **Kit Kat

**Q3: **D

**Honorcode: **1

**Date: **3/15/00

**Time: **10:52:46 PM

The formulas for all of the above tests have a couple of things in common. In the numerator, they all have the difference of two different means. Whether it be the sample and population means (as in the case of the z test) or the means of each population (as in the case of the independent t), each numerator is composed of this difference. The denominators are similar in that they all contain some measure of variance. Whether it be simply the standard deviation (as in the case of the z test) or the more complicated standard deviation of distribution of differences between means (as in the case of the independent t) all denominators are composed of this.

1. To complete a Pearson's r, you would look for words like "relationship", "positive", "negative", and "correlation". Experiments such as these are attempting to find the correlation between two variables. 2. In an independent t, you are looking for two separate and independent populations. One group may have recieved a certain treatment while another group did not. 3. For a dependent t, you are looking for dependent groups. Examples are matched groups such as twins or husbands and wives. You may also look for before/after studies. An example would be blood pressure before and after a certain treatment.

**StudentID: **100984765

**Nickname: **Boomer

**Q3: **D

**Honorcode: **1

**Date: **3/15/00

**Time: **10:57:20 PM

The Z test and One sample T test are similiar in that they are both used when there is only one group. However to differentiate between knowing which one of these test to use for one group problems, the standard deviation is used. If the standard deviation is given, than the Z test should be applied; however, if it is not known, then the One Sample T test should be used. In both the dependent T test and the Independent T test, there are 2 groups. When there are 2 groups, and they are related, then the Dependent T test should be used. In all of the t test, the population variances "are not known and must be estimated."

1. For Pearson's R, I would look for the description incquiring about a correlation between the two groups. Whether positive or negative, it would be looking for a relationship between two variables. Also, I would make sure that the standard deviation was given. 2. For an independent t test, I would look for two groups, which are based on two values of an independent variable. However I would also make sure that the two different groups aren chosen independently of each other. 3. To see if I should use the dependent t test, I would look for two groups, where each one is "paired on a one to one basis, with a single observation in the other sample." For an dependent t test would be used I would make sure that the design depicted that there is a relationship, and that they are not independent unlike the in the independent t test.

**StudentID: **100962404

**Nickname: **Sunny

**Q3: **D

**Honorcode: **1

**Date: **3/15/00

**Time: **11:16:20 PM

The formulas for the a test, one sample t test, dependent t test, and the independent t tests are all geared toward the same basic idea; to derive a formula to find t which consists of the difference of means in the numerator, and standard deviation in the denominator. All the formulas are effected by the number of subjects in the experiment: the more subjects, the higher the probability that we will reject the null hypothesis.

1) In Pearson's r you would look for whether or not the researcher is looking for a relationship between two variables. Each group of numbers in a Pearson's r problem would each represent one, and only one, of the two variables. 2) In an independent t test, the two groups would be independent of one another, and only one dependent variable would be measured at the end of the study. 3) In a dependent t test, each observation in one sample is paired on a one to one basis with a single observation in the other sample. Examples of this are pre and post test, matched samples, and repeated measures.

**StudentID: **101968018

**Nickname: **nicholas

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **12:32:25 PM

All 4 of these tests share the fact that they use a population, for measure or comparison, which is normally distributed.

In the description of the design, in order to find 1) Pearson's r, i would look for a relationship between the two variables, as in a predictable and criterion variable, for 2) i would check the two groups in the experiment are based on two values of one indpenedent variable and that each group is chosen independently of the other, and that one dependent variable is measured at the end of the study; and for 3) a dependent t test i would check to see if each observation in one sample is paired on a one to one basis with a single observation in the other sample.

**StudentID: **101945293

**Nickname: **rynoshaft

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **1:05:52 PM

All three seek to determine a value by dividing the difference of two means over standard deviation. However, T-tests, unlike the Z-test, use an unbiased estimate of standard deviation, which calculated by taking degrees of freedom into account.

1.) In this case you are looking to determine correlation between two variables. An example of this would be to see if people are happier on sunny days than on cloudy days. The r value would show how correlated sunlight is to happiness, but does not necessarily mean that one causes the other.

2.) An independent T test is used to compare two distinct groups, such as people trying to lose weight only through diet vs those who try through both excercise and diet. In this case, the evaluation focuses on how much of an effect, if any, exercise has on weight loss

3.) A dependent T test seeks to compare two separate variables taken from one single group. For example, one may seek to see how length of time on the job relates to job happiness. In this case the variables must be taken from the same people in order to be relevant because they are dependent upon one another.

**StudentID: **101105091

**Nickname: **sly

**Q3: **B

**Honorcode: **1

**Date: **3/16/00

**Time: **1:12:23 PM

the z test, one sample t test, and independent t tests all have only one dependent variable, all have equal interval measurement for hte dependent variable, and al assume a normal distribution.

1) Pearson's r- there are 2 samples where both the mean and standard deviation would be known. 2) Independent t- there are 2 samples where both the mean and standard deviation are unknown, and there are different people in each sample that have been randomly chosen. 3) dependent t- there are 2 samples where both the mean and standard deviations are unknown, and the same paople are in both samples or are matched for a significant reson/similarity.

**StudentID: **101337519

**Nickname: **Trixy

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **3:05:38 PM

All of the formulas for these tests are similar in that they have the same basic form with some slight modifications. The general form is the test statistic equals the difference between the sample mean and the population mean divided by the standard deviation. The formula for finding z, in the z test is z=x-M/SD. The formula for finding t in the one sample t test is t=M-mu/Sm. The formula for t in the dependent t test is t=Md-Mud/(Sd/n^1/2). The formula for the independent t test for finding t is M1-M2/Sdif. one sample t test is t=

1) To use Pearson's r, you would be looking for a relationship between the 2 sets of numbers. If the calculated r is close to -1.0 then there is a strong negative relationship between the two variables in the one sample, and if the calculated r is close to +1.0 then there is a strong positive relationship between the two variables. 2) To use an independent t test the two samples in the experiment must be independent of each other. 3) To use a dependent t test the two samples should be related, for example the repeated measures design, pre-post design, or the matched samples design.

**StudentID: **100936160

**Nickname: **littlebug

**Q3: **C

**Honorcode: **1

**Date: **3/16/00

**Time: **3:14:41 PM

When using a Z-test, the formula is the sample mean - mean of the means/standard deviation: In both the one-sample t-test and the dependent t-test, the formula is the sample mean - population mean/estimate of the standard distribution of means: finally, when using an independent t-test, the formula is the sample 1 mean - sample 2 mean/variances of distributions of differences between means. In all of these equations, we are using some sort of mean to compare two variables. The difference falls in how we are comparing them.

When looking at the description of designs using two sets of numbers, we have to decide which test to use. 1) In the case of using Pearson's r, we are looking for a correlation between two variables. 2) When using an independent t test, we are looking at the differences between mean and the standard deviation of the distribution. 3) For a dependent t test, there will be observations where one sample is paired, on a one to one basis, with a single observation in the other sample.

**StudentID: **100734537

**Nickname: **cheesecake

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **4:25:10 PM

All of these tests involve determining how many standard deviations a particular sample mean is from the mean of the comparison distribution. Both the z test and the one sample t test involve a single group of data from a single sample. In the z test, the variance/standard deviation of the population (comparison distribution) is known, with the one sample t test, this variane is not known, and must be estimated from the sample. While a z test follows a normal distribution, a t test follows a t distribution, which has larger tails and is more spread out than the normal distribution. Because of this, the cutoff score will be more extreme. The cutoff score is also dependent upon the number of degrees of freedom of the sample (n-1). Both the dependet and the independent t tests involve two groups of scores. Dependent t scores are either two scores taken from the same group of subjects (before-after and repeated measures) or from two groups of matched subjects. Independent t scores are from two independently chosen subject groups. The dependent t is similar to the one sample t except that the t score is computed on the basis of the mean difference between scores. The independent t follows a distibution of the difference between means that is also a t distribution.

1) if these data are scores of two different variables given by a single group of subjects and the researcher is asking whether or not there was a linear relationship between the two variables (ex. A=score on a test, B=age) 2) if these data are from two different subject groups that were chosen independently of each other to measure a single variable (ex. measure of introversion for A:North Carolina residents and B:New York residents 3) these data are from either a single group of subjects who were measured on a variable twice (ex. before and after a certain experimental treatment) or are from a set of matched pairs of subjects (ex. members in group 1 are matched with members in group 2 based upon IQ, group one receives treatment, group 2 doesn't)

**StudentID: **100652169

**Nickname: **monkeydoodle

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **4:32:55 PM

In a z-test the standard deviation and the standard error are known while in a a 1-sample t-test the standard deviation and the standard error must be estimated. In a dependent t-test the variances of the population are estimated and in an independent t-test the standard deviation of both populations are given and must be equal. The mean is given for all four tests. Two samples are used in the dependent t-test and the independent t-test while only 1 sample is used with a z-test and a 1-sample t-test.

In the description of the design that would lead you to use a Pearson's r you would look for 2 variables for which there is a relationship but which does not imply causation. An independent t-test is used with 2 groups based on 2 values of one independent variable. Each group is chosen independently of the other. Also, one dependent variable is measured at the end of the study. A dependent t-test is used when each observation in one sample is paired on a one to one basis with a single observation in the other sample. Three approaches can be used including the matched sample design, the repeated measures design, and the pre-post design.

**StudentID: **101954741

**Nickname: **angelfire

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **4:36:56 PM

The z test and the one sample t test, are similar because they both have one group of scores; whereas, the dependent t test and the independent t test have two groups of scores.

To decide which test to use in the problem, we need to ask ourselves a few questions. 1) What do we want to define? If we were looking for a basic relationship or correlation between the two variables, a and b, then we would use the Pearson's r test. If we want a deeper conclusion, such as whether one variable affects the other, we then have to decide between an independent t and a dependent t test. To do this, we have to ask, (2) how are the data sets related? If they are not related, or independent from each other we would use an independent t test. If they are related to each other dependently, we would use a dependent t test.

**StudentID: **101926641

**Nickname: **shakaspara

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **5:00:59 PM

Each test is measuring for the same thing, although the groups of data differ, thus a different test must be used to evaluate the data. Each test is trying to measure where the data gathered lies in respect to the comparison distribution in order to determine whether the results are significant; or to determine whether the null hypothesis should be rejected or retained.

1. If it were necessary to use a Pearson's r to evaluate the data, the description of the design would have to ask me to determine if there was a correlation between the two sets of data. 2. If an independent t test was needed to evaluate the date, the description would state that the two independent groups if individuals had a similar variable, and one group was subject to a treatment. Then the description would ask if the treatment had any effect on the variable the two groups had in common. 3. To utilize a dependent t test, the description would stipulate that the members of group a and group b were the same subjects. The measurements of the variable in group a were taken before the treament, the variables in group b were taken after the treatment.

**StudentID: **101967921

**Nickname: **Betty Boop

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **7:17:00 PM

All of the formulas in these tests have a common theme. The numorator is always a difference between two numbers. In the z-test, one-sample t, and the dependent t, it is a difference between the sample mean and the population mean; and in the independent t test it is a difference between the two samples' means. The denominator in all of these tests is some measure of variance.

If the design had only one group and was looking for a relationship between two variables, that would lead to a Pearson's r test of correlation. If the design had two groups in which these groups were not related to each other, an independent t test would be used. If the design had two groups that were related in some way, either matched pairs, pre-post, or repeated measures, a dependent t test would be used.

**StudentID: **101847140

**Nickname: **tigerlily

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **7:32:37 PM

Z test: A hypothesis testing procedure in which there is a single sample and the population variance is known. Z = (M-mu)/sdm One sample t test: A hypothesis testing procedure in which a sample mean is compared to a known population mean and the population variance is unknown. t = M-mu/Sm Dependent t test: A hypothesis testing procedure in which there are two scores for each participant (or the participants are in matched pairs) and the population variance is unknown: determines significance of hypothesis using difference scores. t = (M-mu)/Sm Independent t test: A hypothesis testing procedure in which there are two seperate groups of people tested and the population variance is unknown. t = M1-M2/Sdif The similarity or commonality between the formulas for these tests is that they all use the population as opposed to a sample.

1) Pearson's r: The average of the cross products of z scores of two variables; a measure of the degree of linear correlation ranging from -1 (perfect negative correlation) through 0 (no correlation) to +1 (perfect positive correlation). What I would look for in the description is how closely the groups are supposed to relate to one another. That is what would be measured.

2) Independent t: Used as a hypothesis testing procedure in which there are two seperate groups of people tested and the population variance is unknown. What I would look for in the description would be the two groups being totally seperate, and having nothing to do with one another. It would have to have both an independent variable as well as a dependent variable.

3) Dependent t: Used as a hypothesis testing procedure in which there are two scores for each participant and the population variance is unknown. What I would look for in the description would be if the two groups are matched together, or paired, if they were in a pre-post study, or if they were in a repeat study. This would show some sort of relationship between the two.

**StudentID: **100959428

**Nickname: **butterfly

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **8:58:00 PM

Each of the equations has on the top the difference between a score, sample or popution mean and on the bottom is the standard deviation or standard error except for the independent t-test, it uses the standard difference. All the formulas give you the test statistic that you need to compare with the cut off value to decide wether or not to reject or accept the null hypothesis.

In a Pearson's r you would look for design in which you are trying to find the average of the cross-products of z-scores, or the degree of linear correlation. In an Independent t test you would look for a design comparing the means of two entirely seperate groups of people whose scores are independent of each other. In a dependent t test you would look for a design in which you have scores for a single sample, you use difference scores(for each person you subtract one score from the other) and you want to compare this to a population for which you know the mean( you assume that it is 0), but not the variance.

**StudentID: **101169420

**Nickname: **quackers

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **8:58:15 PM

Each of the tests are and equation that consists of on the top a difference b/w a score, sample, and popultion mean. On the bottomis the standard deviation or standard error, except the independent t-test which uses the standard difference. Each of these tests give you the test statistic need to compare with the given cut-off value, which allows you to decide whether or not to accept or reject the null hypothesis.

1: you would look for a design where you try to find the average of the cross-productsof the z-scores or the linear correlatioon degree. 2: you find a design that compares the means of two separate groups of subjects, whos scores are independent of the other groups. 3: the design is where you have all of the scores for a single sample and use the differences ofeach persons scoreto coompare to a population that you know the mean for, which you assume is equal to zero.

**StudentID: **101176381

**Nickname: **scooter

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **9:03:48 PM

All the formulas are looking for a value with which to compare to a cutoff value for each test to determine if the null hypothesis will be rejected. All od the formulas use some form of mean of means in comparison with the standard devaition or expected SD.

1) You would use a Pearson's r if the A and B were representative of two varibles which were being compared. In other words, you would use Pearson's r if you were examining the relationship between the two variables, A and B. 2) You would use an independent t test if A and B were two separate groups of people in a test, and you wanted to compare the results of the two independent groups. 3) You would use a dependent t test to evaluate the data if A and B were paired samples in a study and you wanted to compare them. Or if they were the same people, but in a before and after situation(i.e. paired with themselves before and after).

**StudentID: **100982563

**Nickname: **Yoda

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **9:09:51 PM

First of all, for all these tests you have to set your alpha level or confidence interval to determine how sure you are in your findings, we tend to choose either alpha levels of .05 or point .01. Secondly, you have to determine your cutof scores from the charts in the back of the book for all these tests to put in their respective formulas, whether it is a t test or a z test; one sample t tests, dependent t test and independent t test all use the same table. All 4 formulas require, first of all, the mean. For the z test and the 1-sample t test we use the the mean of the population designated by the greek letter mu. All tests are based on the assumption that there is a normal distribution in the sample or population the numbers were taken from. All formulas use a form of sigma to determine the respective scores. However, for t tests, sigma is tweeked a little bit because sigma is always too low. So for 1-sample t tests we take SS and divide it by the df and then take the square root to find a more unbiast standard deviation. For the dependent t, the SD is also necessary but it is found by the average of D, which is X1-X2 for all pairs. The independent t test uses the pooled variance from the two different variables. Once we have obtained our respective z or t scores, in all formulas you compare that score to the signifcance cutoff point found in the chart to determine whether or not the results of your research is significant or not. The number of subjects is also important to all tests and their formulas because increasing the number of subjects increases the power of the overall experiment. To make one thing clear about sigma of m, for a z test simga is known from a normal distribution or large sample size while in a t test it is not known so it must be estimated.

From looking at the above data, I believe I would conduct an dependent t test to evaluate the data, that is bending of course that this data is the result of a pre-post test which it appears to be. There is the same number of subjects in A and B and it appears that a test was taken before a treatment was administered, then treatment was given, and then the test was given again to determine the effects of the treatment. From the results, it would appear that the test reduced whatever was treated, but to determine the significance of the experiment you would have to conduct a dependent t test.

**StudentID: **101194842

**Nickname: **moggy

**Q3: **C

**Honorcode: **1

**Date: **3/16/00

**Time: **9:29:44 PM

All four of these populations are normally distributed. The z-test and the one sample t-test only have one sample. In the independent, dependent and one sample t-tests the variance is not known.

1)In a Pearson's correlation coefficient is used when you want to measure the degree of linear correlation ranging from -1 to +1. You would use a z-score, which is an inferential statistic and the data would have to have only one sample. The mean and standard deviation of the comparison is known. 2)You would use an independent t-test when there are two seperate groups of people tested and the population variance is not known. 3)You would use a dependent t-test when there are two scores for each participant or in other words they are in matched pairs. The population variance is unknown.

**StudentID: **100879061

**Nickname: **brianimal

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **9:31:54 PM

One similarity between the formulas of these for tests is that they all use the value of the sample mean. In the z test, one sample t test and the dependent t test, the formula all has the difference between the sample mean and the mean of an entire population in the numerator. In the independent t test, it's the mean of one sample minus the mean of another sample in the numerator. Another similarity between the four tests is that all of the formuals use some sort of standard error in the denomenator.

In using Pearson's r, you would look for whether there is one group or level and then look to see if there is a relationship between the 2 variables. In using an independent t test, you would look to see if there were two groups or levels and also look to see if those two groups are independent of each other. In using a dependent t test, you would look to see if there are two groups or levels and also to see if those two groups are related to each other.

**StudentID: **100996799

**Nickname: **tank

**Q3: **B

**Honorcode: **1

**Date: **3/16/00

**Time: **9:32:09 PM

all these formulas have in common they use the standard deviation, the means, to see if it is one tailed or a two tailed test. they also to see if there is any difference in the signifcant level. Also you can reject the null hyppthesis with these formulas from the answers you get from them.

You would use the independent to evaluate the data.

**StudentID: **56695

**Nickname: **Figure8

**Q3: **A

**Honorcode: **1

**Date: **3/16/00

**Time: **9:46:15 PM

Each formula is conducive to obtaining useful, significant information relating to hypothesis testing.

I would like to see; a) a cause and effect relationship or b) paired samples/related groups or c) two samples that are entirely seperate from each other.

**StudentID: **101905648

**Nickname: **rygy

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **10:02:14 PM

The formulas for z test, one sample t test, dependent t test, and the independent t test all have things in common. The formulas are as follows; the ? z test- t= (M-um)/om ? one sample – t= M-u/Sm ? dependent – t= M-u/Sm ? independent – t=M1- M2/Sdiff

After looking at all these formulas you can see that they all really are the same. Each of the four formulas take the mean of the sample and subtract it from either the mean of the population or the mean of the other sample (independent t test). Then they all divide that total by some form of the standard deviation.

1. When looking for Pearson’s r you would look for things in the question that ask you to determine what type of, if any, relationship exists between the two variables. When asked to determine a relationship you know you should be doing a correlation (Pearson’s r) 2. When looking for an independent t test you would look for things in the question that determine there to be two different groups within the same sample to test things against each other. If you see this then you know that you should be conducting an independent t test. 3. When looking for a dependent t test you would look for things in the question that refer to one group that is observed at two different times during the experiment. If can either be a match pairs, pre-post design, and also repeated measures design. Any of these three will tell you that you are supposed to be doing a dependent t test.

**StudentID: **101816052

**Nickname: **chou-chou

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **10:41:15 PM

In finding the values for all 4 of these tests, the mean of the population (or other sample in case of dependant and independent t-tests) is subtracted from the sample in the numerator of the equation. For example, for z test, z=(M-mean of pop.)/standard deviation. And for the t test, t=(M-mean of pop.)/(s/sq. root of n). For all of the t tests, however, the degrees of freedom matter in calculations.

1) For a Pearson's r, the description would probably include a curvilinear relationship between two variables. This means that by increasing or decreasing one of the variables, the other one either increases or decreases. Also, if you could draw a scatter diagram between the two variables. 2) For an independant t test there are two groups in an experiment based on two values of an individual variable and we assume that the variances and means for both groups are equal. We also would assume that 2 random samples, the 2 raw score populations are normally distributed and standard deviations of each are equal. 3) For a dependant t test the two groups are related to each other in some way. This could be by matching, a pre-post study, or repeated measure design. Each observation from one group is paires with one from the other group in a 1-to-1basis.

**StudentID: **101974684

**Nickname: **buttercup

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **10:53:47 PM

All four formulas require the differences of the two means in the study (either the combination of population and sample means or the combination of two population means) as the numerator in the equations. The z-test, one sample t-test, and dependent t-test all use the difference between the sample mean and the population mean. For the independent t-test, the difference between two different population means is used. The denominators in the four equations are all variations of the standard deviations. For instance, the z-test formula requires the population standard deviation, while the one sample t-test requires an estimated standard deviation of the distribution of means.

1) To use Pearson’s Correlation Coefficient (r), one would be trying to find a relationship between two variables, such as GPA and hours spent studying. The question would be presented in a manner such as, “Do students who study more have higher GPAs?” A hypothesis test would be run, using critical values for Pearson’s r and the computation of the r according to the problem. 2) When using an independent t, one would look for two separate groups "based on two values of an independent variable." The two groups would be chosen independently of one another. Also, one dependent variable would be measured at the end of the study. This is a situation where a t-test is used because “the population variances are not known and must be estimated.” 3) A dependent t requires two groups of scores for the same groups of people, on the other hand. These groups are related to each other instead of being independent of one another since these scores come from the same participants. The scores in this situation are before and after scores.

**StudentID: **101175223

**Nickname: **rodimus

**Q3: **C

**Honorcode: **1

**Date: **3/16/00

**Time: **11:03:50 PM

each uses the score minus the mean and divide by the standard deviation. in some cases the standard deviation is divided by the square root of n.

for pearson's r we must look for a relationship between 2 variables within one group. for the independent t test we need two independent groups. the dependent t test requires 2 related groups of some sort.

**StudentID: **100747476

**Nickname: **goat

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **11:07:33 PM

The formulas for each mentioned test have many similarities, as well as differences. To begin, each formula holds the common goal of testing whether or not to reject a null hypothesis. Each formula uses the difference of means, whether they are of a sample, population, or mean of means. Each formula has some combination of differences of means as its numerator. Additionally, the denominator of each formulas consists of a standard deviaton. These are sometimes know, and sometimes calculated or estimated, but it is this standards deviation that makes up the denominator of each formulas. Another similarity among formulas is that each assumes a normally distributed population. Overall, each formula is a means by which to attain a common goal: a conclusion of a hypothesis test. The formulas use similar figures (ie: mean, st. dev., etc.) The difference is where those figures come from (ie: sample, population, estimation, etc.)

The description that would suggest the use of a Pearson's 'r' would be an indication of looking for a correlation. Meaning, if the description showed a search for a relationship (positive or negative) between the two sets of data, then one could concluded that Pearson's r was the correct test. If it were meant to be a test for an independent t, there must be a statement declaring that the two sets of data are not related. Additionally, the standard deviations of each population is equal and must be estimated. Therefore, the key result would come from the differences of means of each sample. In order to imply a dependent t test, it must be stated that each set of data is somehow related to the other. (ie: 2 different types of scores per 1 person.) Both the population mean and variance would be unknown.

**StudentID: **100951026

**Nickname: **shera

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **11:12:09 PM

z test- used in experiments with ONE sample. you must know the mean and the standard deviation of the comparison distribution. one sample T-test: ONE sample, mean of population is known, but standard deviation of population is not known. Dependent T-test : each observation in one sample is paired one a one - to one basis with a single observation in another sample. ( matched, repeated measures, pre-post) Variances must be estimated and variances are assumed to be equal. you must compute average differences. Independent T-test : have two random samples. The variance of each population is equal and must be estimated from the stand. Dev of each sample

Z-test and one sample T-test are similar because they are both used with ONE sample. They are also set up similarly in that they both involve the difference of means in the numerators and divide by standard errors in the denominators. But a difference is that in the Z test the standard deviation of the comparison distribution is given, and in the one sample T-test the standard deviation of the population is not known and therefore must be estimated using an unbiased method: S= the square root of ( SS/n-1). The dependent T-test is similar to the one sample T-test because the variance has to be estimated as well, and in the denominator is the standard deviation divided by the square root of n. The difference of the Dependent T -test is that it uses the average differences instead of the difference of means in the numerator. The independent T-test is the most different of all the tests because it is the differences of the sample means and then divided by just the standard deviation of the differences.

For a 1) Pearson's r, I would be looking for a design that involved a relationship or correlation between two groups. I would look for an experiement where one is trying to prove that group A is related to B, either with a positive correlation or a negative one, or it could be that the two groups are just different. For a 2) independent t test- i would look for a design with two groups based on two values of one independent variable. Each group would be chosen independently of the other, and then one dependent variable would be measured at the end of the study. For a 3)dependent t-test- I would look for pairing, matching, repeated measures design or pre-post design.

**StudentID: **101852531

**Nickname: **Daisy Girl

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **11:12:26 PM

The formulas for variance of each of the four are all summations of the mean minus raw scores divided by the number of subjects or degrees of freedom, with only slight variations in each. The formulas for the actual tests are all forms of the differences between two means, divided by some form of the standard deviation.

1( Pearson's r: I would look to see if the problem was a correlation. Data A and data B would have to be to measurements of two different things and the question would be asking to determine if there was a correlation between the two. For example is A were points correct on a test and B was hours spent studying for that test, and the question was whether or not to students who study more will recieve higher grades on the test. 2) an independent t test - If the design of the experiment was two independent groups, each cbosen to with no relation to the other, and only one variable is measured at the end of the study. For Example. Two groups of obese people are chosen; one group recieves a drug to cause weight lose, while the other is given no treatment at all. Research is done to determine if people given the drug will lose more weight then those who do not recieve the drug. 3)dependent t test: This test is used if the experiment is run with one sample of subjects and the Standard Deviation of the Means is unknown and must me estimated using the unbiased sample SD. Only one dependent measure is taken from each participant. An example of this is: the average burnout score of college seniors is 8.9 on a scale of one to ten. Is the average score of a ample of 10 Psyc Seniors less than the population of seniors.

**StudentID: **101264596

**Nickname: **droopy

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **11:20:21 PM

all of the formulas have a population mean being subtracted from a sample mean and they are all divided by a standard deviation or a n estimate of a standard deviation

with Pearson's r you would have to have a relationship between the two variables; independent t would have two independent groups; a dependent t test would have two related groups of numbers

**StudentID: **100986806

**Nickname: **suzyq

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **11:22:11 PM

Assuming we use the Z-test for means, all of the tests must have an "N" larger than 1, the number of levels for each independent variable is 1, the level of measurement is equal interval, the number of dependent variables is 1, and we assume a normal distribution. More specifically, the Z test for means and t-test for one sample have a known population mean. The t-test for dependent and independent means has an unknown population mean, unknown population variance, and an unknown population standard deviation.

1)Pearson's r--Look for 2 independent variables on an equal interval scale. You would also look for anything that discusses a correlation between the samples. 2) independent t--look for unknown pop mean, unknown pop variance and SD, N larger than 1, different group of people in each sample. 3) dependent t-- unknown pop mean, unknown pop variance and SD, matched sample, N larger than 1.

**StudentID: **100508589

**Nickname: **squid41

**Q3: **C

**Honorcode: **1

**Date: **3/16/00

**Time: **11:26:36 PM

They all have to do with means and use the Standard devation of the means in the equation along with the mean of the sample or population

I would look for weather it was comparing something similiar to a cause and effect relationship even though correlation is not a causal relationship it just shows a relationship, if two populations were being tested to see weather or not they are independent from one another

**StudentID: **100710111

**Nickname: **goose

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **11:39:08 PM

To begin, each test is designed to test a hypothesis. For each of the t-tests, the population variance is not known; for the z-test, the population variance is known. With the z-test and one sample t-test, only one sample is used. The dependent t-test allows for one sample to be tested twice, once before and once after "treatment". The independent t-test has two samples totally different from the other. For both the z-test and one sample t-test, the mean is given. Also the z-test is the only test where the assumption of the normal distribution is not needed.

You would expect to see the mean and population variance of both sets of data to use the Pearson's r. You would also expect there to be only one sample. If it were a independent t, the design should state that the each set of numbers have nothing to do with each other and the design would be based on two values of an independent variable with one dependent variable measured at the end. With a dependent t-test, the design would be say these numbers from group A were before the treatment and numbers from group B were after the treatment and you would try to find a significance in the changes from each group. The groups would come from the same sample, just one group before the affect and one after.

**StudentID: **100896105

**Nickname: **honeydo

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **11:41:52 PM

The z-test and the one sample t test both use one sample group. What makes them different is that with the z-test the standard deviation is given and with the one sample t-test, the standard deviation is not given. THe independent t and the dependent t bot use two groups. What differs between the dependent t and the independent t is that the dependent t uses two groups that are related (ie: by marriage, IQ, age, etc...) whereas the independent t uses two groups that are independent or unrelated to each other.

- if you needed to find Pearson's r, the description would ask for the relationship between variable A and variable B

- if you needed to use an independent t test, the description would ask say that the Group A and Group B are not related

- if you needed to use a dependent t test, the description would say that the Group A and Group B were related to each other in some way.

**StudentID: **101250465

**Nickname: **Pebbles

**Q3: **C

**Honorcode: **1

**Date: **3/16/00

**Time: **11:43:37 PM

These four formulas have their basic structure in common: (Population mean - Sample Mean)/ Standard Deviation of some set of numbers.

If the description of the design said something to reveal that they were looking for a correlation among the variables then you would use the Pearson's r. If the description of the design revealed that the experiment was based on two values of an independent variable where each group is chosen independently of the other then you would use the independent t test. If the description of the design leans towards one of the three tests such as matched samples, repeated measures, or pre-post test then you would use the dependent t test.

**StudentID: **101278100

**Nickname: **skibum

**Q3: **C

**Honorcode: **2

**Date: **3/16/00

**Time: **11:49:50 PM

one common thing between the 4 formulas is that the use some sort of form of a mean and then some sort of a standard deviatopn. these are either given to you in the problem or they are found. here are the formulas. Z test z=x-m/sd ; One Sample t Test = t = M - mu/Sm ; Dependent t test = t - Md-muD/(Sd/ sq rt n) and finally Independent t test = t = M1-M2 / Sdif

To see if you were looking for a pearsons r you may look and see if you were looking for any correllation between the two number sets. To see if you were looking for a dependnet t test you would look and see if if the numbers were paired together or if they were repeated or pre and post experiment. in doing this you would be looking to see if one of the set of numbers was dependent on the other and finally to see if you were looking for an independent t test you would look to see if the numbers were totally and completly seperate from each other and that they were in no way related at all.

**StudentID: **100881989

**Nickname: **scrubski

**Q3: **D

**Honorcode: **1

**Date: **3/16/00

**Time: **11:51:08 PM

All of these tests have the difference of two means in the numerator. They are also divided by some sort of standard deviation, whether it be estimated or a known standard deviation.

For Pearson's r there would have to be only one group, it measures the relationship between two variables. So if A and B were IQ and creativity, I would use Pearsons r. To use a dependent t-test there would have to be two related groups and it is based on the differences in the two groups. AN example is if A measures abstract logic before a special course and B measures arstract logic after the course. An independent t-test it is the comparison of two seperate groups whose scores are independent from one another. If A and B were both groups of obese males on two different excercise programs, and I used weight loss to measure the effectiveness of these programs, I would use the independent t-test.

**StudentID: **100632956

**Nickname: **froggy

**Q3: **D

**Honorcode: **1

**Date: **3/17/00

**Time: **12:18:54 AM

For all the tests there is always a subtraction of means which is then divided by some form of standard deviation.

For a Pearson's r test a correlation would have to be asked for in the design. This means the experimenters are trying to see if there is a strong relationship between the two variables. For an independent t test, A and B would have to be data from two different groups and the purpose would be to find the difference in means. For a dependent t test A and B could be paired scores.

**StudentID: **101921521

**Nickname: **sweets

**Q3: **D

**Honorcode: **1

**Date: **3/17/00

**Time: **12:24:03 AM

For all four formulas, the thing that they have in common is the fact that they are each the difference of two means in the numerator, whether it is a sample or population. Then, this difference is divided by a standard deviation that is either estimated or known. Here are the formulas for each example:

Ztest: 1: An inferential Statistical test 2: use in experiments with one sample 3: must know the M and O of the comparison distribution 4: involves the use of the distribution of means M-Mm / Om One sample test: 1: Random smaple of scores 2: Population from which the sample is drawn is a normal distribution 3: one dependent measure is taken from each participant t = M-M / Sm Dependent test: 1: Each observation in one sample is paired, on a one to one basis with a single observation in the other sample t = Md-Md / Sd/square root of n Independent test: 1: Two groups in the experiment are based on two values of an independent variable: each group chosen independently of the other 2: One dependent variable measured at end of study t = M1 - M2 / Sdif

Pearson's R: This descriptive statistic describes the degree and direction of a linear correlation in the particular group of people studied. r = the sum of (Zx)(Zy) / N ( average of the cross products of Z scores) - you measure two different variables for one sample - draw a scatter diagram to show clear curvilinear pattern - an example problem would be : A researcher is interested in the relation between psychotherapists' degree of empathy and their patients' satisfaction with therapy. ( the patients and their therapists are paired into one sample)

Independent t: Two groups in experiment based on two values of an independent variable. - Each group chosen independently of the other - One dependent variable measured at end of study - Two groups are each given a different value of an independent variable t = M1 - M2 / Sdif - An example problem would be : Two groups of males are chosen to see if weight reduction in Program 1 causes a better weight reduction than in Program 2.

Dependent t : Each observation in one sample is paired, on a one to one basis, with a single observation in the other sample. - Three types: Matched Sample Design Repeated Measures Design Prepost Design t = Md - Md / Sd / Square root of n - an example of a problem: A program to decrease littering was implemented in four cities in California's Central Valley starting in August 1997. The amount of litter in the streets was measured during the July before the program started and then the next July, after the program had been in effect for a year. (Prepost Design)

**StudentID: **101137818

**Nickname: **giggle box

**Q3: **D

**Honorcode: **1

**Date: **3/17/00

**Time: **12:30:09 AM

They all result in the same thing by telling if the null hypothesis about a certain sample was true or not. They all include knowing the mean of a certain problem in order to calculate the value.

1) when only 1 sample is used 2) 2 related groups are compared 3) 2 independent groups are compared

**StudentID: **101435626

**Nickname: **Vokamis

**Q3: **D

**Honorcode: **1

**Date: **3/17/00

**Time: **12:41:06 AM

They all attempt to see how far above or below a certain score the sample scores are.

1) Relationship in the degree of the two sets. 2) Difference between the means of the two unrelated (different) sets. 3) Difference between the means of the two related (same) sets of numbers.

**StudentID: **101125631

**Nickname: **monkeygilr

**Q3: **D

**Honorcode: **1

**Date: **3/17/00

**Time: **12:46:13 AM

You are always dividing the difference of means by the standard error.

1) Look for a comparrson or relationship between two variables. 2) 2 groups in experiment based on two values of an independent variable. Each group is chosen independently of the other. One dependent variable is beign measured at the end of the study. 3) Look to see if it is a matched samples design, repeated measures design, or a pre-post design.

**StudentID: **101244082

**Nickname: **apple bum

**Q3: **D

**Honorcode: **1

**Date: **3/17/00

**Time: **12:53:53 AM

All the levels of measurement for the dependent variables are equal interval. We assume normal distribution for the distribution of the dependent variable. There is only ONE independent variable. The number of people in the samples is always greater than one. All the comparison distributions are forms of distributions of means: Z-test--distribution of means, one sample t-test--distribution of means, dependent t-test--distribution of means of difference scores, independent t-test--distribution of differences between means.

1) In order to use Pearson's r, we would look to make sure that there are two variables to measure, both on equal-interval scales. The null hypothesis must assume that rho=0, meaning no relation. 2) In an independent t test, we would look to check that we don't know the values of the population mean nor variance nor standard deviation, and we also assume normal distribution. We also need to find out how the people were assigned to groups, where need to have been randomly assigned with different people in each group. 3) For the dependent t-test to be performed, we need to check that there are unknowns: population mean, population variance, and population standard deviation. We also need to know the assignment of people in the group was matched, with the same peoples being in each group w/ matched scores. We assume normal distribution and that the measurements are equal interval.

**StudentID: **101099628

**Nickname: **smurf

**Q3: **D

**Honorcode: **1

**Date: **3/17/00

**Time: **1:06:16 AM

The numerator in the formula is always the difference between the two means. Predicted mean and the comparison mean.

If I were to use a Pearson's r test to evaluate this data there would have to be a question of whether there is a relationship between the to data set. df is n-2 and p=0. If I were to use an independent t test to evaluate this data the two groups in the experiment would be based on two values of an independent variable, and each group would be chosen independent of each other. If I were to use a dependent t test to evaluate this data, the two groups in the experiment would be based on two values of dependent variable.

**StudentID: **101220954

**Nickname: **Nash

**Q3: **D

**Honorcode: **1

**Date: **3/17/00

**Time: **1:23:54 AM

The basic principle of hypothesis testing is used in each test where you are comparing two samples, such as an experimental group and a control group. Also you are comparing the means in the formula where you determine the t or z for the experiment. M1-M2 is at the top of the formula in each test.

The first thing that you would look for in the set of numbers design would be the number of groups in the set of numbers. If it had 2 groups of numbers, you would use either a dependent or independent t test. If it had one group fo numbers you would use either a z test, a one sample t test depending on whether or not you were given the standard deviation, or Pearson's r . In this case it would be either a dependent or independent t test. The next thing you would look for would be if the two set of numbers were independent of eachother or they were related. If they were related, you did'nt know the variance, and in the same group of people, you would use a dependent t test. If they were independent of eachother, the variance still unknown, and groups were not relying on eachother, or related, you would use an independent t test. In these this set of numbers we do not know if A and B are related, so it could be either a independent or a dependent t test.

**StudentID: **101229881

**Nickname: **Artee

**Q3: **D

**Honorcode: **1

**Date: **3/17/00

**Time: **2:26:30 AM

??

When given two sets of numbers for an experiment, you would look for different descriptions in design which would lead you to use different tests to evaluate the data. In order to assume that you would need an independent t test, there would have to be two groups and the group would have to be independent rather than related. They also would be interval or ratio scores and the two raw score populations would be normally distributed. To use the dependent t test, the data would have to have two groups but these groups would have to be related somehow. For example, maybe there are two groups of scores recorded but for the same people. The Pearson's r is used when there is one group given and there is a relationship between the two variables.

**StudentID: **101835826

**Nickname: **Lightning Bolt

**Q3: **D

**Honorcode: **1

**Date: **3/17/00

**Time: **2:39:54 AM

The similarity between the four formulas is that all formulas are trying to find the standard deviation of the distribution of the means which then will indicate whether the research or hypothesis they have is true or false.

1) Pearson's r : An example of this kind of problem would be if these numbers were used to describe the relationship between the two variables. You are looking to see if X is causing Y or Y is causing X or a third factor is causing both X and Y. If so, you need to run a hypothesis test to see if there is a postive or negative relationship between the two variables. 2) Independent t-test: Use this test if you are looking for the difference between the means of two samples. Two populations are being considered. An example would be one population that an experimental group is taken and one polulation that a control group is taken from. A researcher will not know the mean of either population, therefore the results from the findings and comparisons of the means will determine of the null hypothesis is true or not. 3) Dependent t-test : This test is done where each participant has two scores, such as before a score and after score. You first have to figure a difference score for each participant then you carry out a hypothesis test because the variance of the population is unknown. The key is that you have to estimate the population variance from the scores in the sample (use formula: divide sum of squared deviations scores by the degrees of freedom).

**StudentID: **101274025

**Nickname: **SA

**Q3: **C

**Honorcode: **1

**Date: **3/17/00

**Time: **3:29:06 AM

In all of these tests, (z test, one sample t test, dependent t test and independent t test) the formulas for computing either the z or the t are the same thought. In the numerator, you always take the difference of the means. In certain tests it is the difference of 2 different sample means and in others it is the population mean and the sample mean. The denominator is always related to the variance. In some it is devided by the number of subjects being used and other tests simply use the standard deviation. It is all following the same premise.

a) Pearson's r is used in correlations. So, I would look for the design to be testing to see if two separate entities were related in any way or had an effect on one another. By calculating Pearson's coefficient, I can see how stronly or weakly sets A and B are related to one another. b) The independent t test requires a few things. First, I would look to see that there were 2 random samples or interval or ratio scores. There would have to be 2 groups in the experiment based on 2 values of an independent variable with each group chosen independent of the other. At the end, I would evaluate the data by measuring one dependent variable. c) For a dependent t test, I would be looking or information within the design that would tell me if scores from A and B were either matched sample designs, repeated measures design or a pre-post design. By knowing this, then I know that the scores from A are eache paired with a score from B. I would then know that I need the dependent t test in order to evaluate the data.

**StudentID: **101142423

**Nickname: **mighty mouse

**Q3: **D

**Honorcode: **1

**Date: **3/17/00

**Time: **6:55:05 AM

With a one sample t test and a dependent t test, S squared is equal to the summation of of either the differences or x minus the mean of either subscript x or d, and then you square that, with all of that divided by N-1. For the one sample t test, S squared=summattion (X-Mx)squared/N-1. For the dependent t-test, S squared=summation (D-Md)squared/N-1. They all deal with subtracting the mean from X. T

To find Pearson's r, you would have to find the mean and the standard deviation. You also have to see that there are two different, independent variables. With a dependent t test, you would look to see that the variance is not described. You also look to see that the two sets of data are dependent of each other. You would look to see if it was a matched samples design, a repeated measures design, or a pre/post design. With the independent t test, you would look to see that the two groups are both independent of each other. You look to see that you have two random samples.

**StudentID: **100689400

**Nickname: **Chip Douglas

**Q3: **D

**Honorcode: **1

**Date: **3/17/00

**Time: **7:57:38 AM

The formulas for the effect sizes for each of the test is the same, M1-M2/Population Standard deviation. The formula to find the t score and the z score are the same, except that you find a t score based on a t distribution rather than on a normal distribution which characterizes a z score. Also, the t score for independent means differs from the t score for dependent means and the z score, in that you calculate it on a distribution of differences betwwen means rather than on a distribution of means.

To use Pearson's r you look for a correlation or relationship between the two variables. Using Pearson's r, correlation is determined by calculating the average of the cross products of z scores of two variables. However, it is based on a single sample and in this case ther are two samples. If you were to use a t test for dependent means you would look to see if there is a single group of participants, in which there are two scores for each participant. To use a t test for independent means you would look to see that there are two separate sample groups being tested. Thus, in this design, you would use the t test for independent means.

**StudentID: **100833872

**Nickname: **hairball

**Q3: **D

**Honorcode: **1

**Date: **3/17/00

**Time: **9:29:00 AM

Population means, population variance, and standard deviations are all common elements to formulas for the Z test.

1.pearson's r=look for and independent variable and a dependent variable in order to show the degree of linear correlation. 2.dependent= the two populations are related to each other or somehow connected 3independent= the two populations are not connected at all