This document introduces the t-statistic as an alternative to the z-score for hypothesis testing when the population standard deviation is unknown. It explains that the t-statistic uses the sample standard deviation to estimate the standard error. It provides examples of when to use a t-test and steps for conducting a t-test, including computing the test statistic, determining the degrees of freedom, and referencing the t-distribution table. The document also gives an example problem demonstrating how to perform a hypothesis test using the t-statistic to compare two sample means.
This document introduces the t-statistic as an alternative to the z-score for hypothesis testing when the population standard deviation is unknown. It explains that the t-statistic uses the sample standard deviation to estimate the standard error. It provides examples of when to use a t-test and steps for conducting a t-test, including computing the test statistic, determining the degrees of freedom, and referencing the t-distribution table. The document also gives an example problem demonstrating how to perform a hypothesis test using the t-statistic to compare two sample means.
Original Description:
T-statistics, Psychological Statistics, college notes, handout from professor.
This document introduces the t-statistic as an alternative to the z-score for hypothesis testing when the population standard deviation is unknown. It explains that the t-statistic uses the sample standard deviation to estimate the standard error. It provides examples of when to use a t-test and steps for conducting a t-test, including computing the test statistic, determining the degrees of freedom, and referencing the t-distribution table. The document also gives an example problem demonstrating how to perform a hypothesis test using the t-statistic to compare two sample means.
This document introduces the t-statistic as an alternative to the z-score for hypothesis testing when the population standard deviation is unknown. It explains that the t-statistic uses the sample standard deviation to estimate the standard error. It provides examples of when to use a t-test and steps for conducting a t-test, including computing the test statistic, determining the degrees of freedom, and referencing the t-distribution table. The document also gives an example problem demonstrating how to perform a hypothesis test using the t-statistic to compare two sample means.
Copyright:
Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online from Scribd
Download as pdf or txt
You are on page 1/ 7
Usapang t
INTRODUCTION TO THE t STATISTIC
THE T STATISTIC: AN ALTERNATIVE TO Z
Psych 110 Day 13
The t statistic: An alternative to z
z score: requires more info than is usually available (e.g. population SD or variance to compute the SE) Usually, the SD for population is unknown. When the population SD (or variance ) is unknown, we use the t statistic instead of a zscore for hypothesis testing.
The t statistic: An alternative to z
Thus, the t statistic can be used to test hypotheses about a completely unknown population; that is, both and are unknown, and the only available information about the population comes from the sample. All that is required for a hypothesis test with t is a sample and a reasonable hypothesis about the population mean.
Two general situations where this type of hypothesis test is used
1.The t statistic is used when a researcher wants to determine whether or not a treatment causes a change in a population mean. In this case you must know the value of for the original, untreated population. A sample is obtained from the population and the treatment is administered to the sample. If the resulting sample mean is significantly different from the original population mean, you can conclude that the treatment has a significant effect.
Two general situations where this type of hypothesis test is used 2. Occasionally, a theory or other prediction will provide a hypothesized value for an unknown population mean. A sample is then obtained from the population and the t statistic is used to compare the actual sample mean with the hypothesized population mean. A significant difference indicates that the hypothesized value for should be rejected.
The Estimated Standard Error and the t Statistic
Whenever a sample is obtained from a population you expect to find some discrepancy or "error" between the sample mean and the population mean. This general phenomenon is known as sampling error. The goal for a hypothesis test is to evaluate the significance of the observed discrepancy between a sample mean and the population mean.
The Estimated Standard Error and the t Statistic (cont.)
The hypothesis test attempts to decide between the following two alternatives: 1. Is it reasonable that the discrepancy between M and is simply due to sampling error and not the result of a treatment effect? 2. Is the discrepancy between M and more than would be expected by sampling error alone? That is, is the sample mean significantly different from the population mean?
The Estimated Standard Error and the t Statistic (cont.)
The critical first step for the t statistic hypothesis test is to calculate exactly how much difference between M and is reasonable to expect. However, because the population standard deviation is unknown, it is impossible to compute the standard error of M as we did with z-scores. Therefore, the t statistic requires that you use the sample data to compute an estimated standard error of M.
The Estimated Standard Error and the t Statistic (cont.)
Estimated standard error (SM) used as an estimate of the real SE (M)when the value of is unknown. It is computed from the sample variance or sample SD and provides an estimate of the standard distance between a sample mean M and the population mean .
The Estimated Standard Error and the t Statistic (cont.)
We must use the sample variance, s2, in place of the unknown population variance, 2 (or use sample standard deviation, s, in place of the unknown population standard deviation, ). The resulting formula for estimated standard error is SM = or SM = We will now use the variance in our formula for standard error.
The Estimated Standard Error and the t Statistic (cont.)
The t statistic (like the z-score) forms a ratio. The top of the ratio contains the obtained difference between the sample mean and the hypothesized population mean. The bottom of the ratio is the standard error which measures how much difference is expected by chance. obtained difference M t = = standard error sM
The Estimated Standard Error and the t Statistic (cont.)
A large value for t (a large ratio) indicates that the obtained difference between the data and the hypothesis is greater than would be expected if the treatment has no effect.
The t Distributions and Degrees of Freedom
You can think of the t statistic as an "estimated zscore." The estimation comes from the fact that we are using the sample variance to estimate the unknown population variance. With a large sample, the estimation is very good and the t statistic will be very similar to a z-score. With small samples, however, the t statistic will provide a relatively poor estimate of z.
The t Distributions and Degrees of Freedom (cont.)
Degrees of freedom (df) the number of scores in a sample that are independent and free to vary. Because the mean places a restriction on the value of one score in the sample, there are n1 df for the sample.
The t Distributions and Degrees of Freedom (cont.)
The value of degrees of freedom, df = n - 1, is used to describe how well the t statistic represents a zscore. Also, the value of df will determine how well the distribution of t approximates a normal distribution. For large values of df, the t distribution will be nearly normal, but with small values for df, the t distribution will be flatter and more spread out than a normal distribution.
The t Distributions and Degrees of Freedom (cont.)
To evaluate the t statistic from a hypothesis test, you must select an level, find the value of df for the t statistic, and consult the t distribution table. If the obtained t statistic is larger than the critical value from the table, you can reject the null hypothesis. In this case, you have demonstrated that the obtained difference between the data and the hypothesis (numerator of the ratio) is significantly larger than the difference that would be expected if there was no treatment effect (the standard error in the denominator).
The t distribution The t distribution is an approximation of the z distribution. To evaluate a t statistic, the critical region must be located in a t distribution. Family of t distributions: exact shape depends on df. Thus, the critical t values will depend on the value for df associated with the t test. As df increases, the shape of the t distribution approaches a normal dist.
Using the t-distribution table
Two rows at the top: proportions of the t distribution contained either in one or two tails, depending on which row is used. First column: df of the t statistic Note that as df increases, the t distribution becomes more similar to a normal dist. When your t statistic has a df value that is not listed in the table, look up the critical t for both of the surrounding df values and then use the larger value for t.
Using the t-distribution table
How to compute for the t statistic
1. Calculate the sample variance (or sample SD) as a substitute for the unknown population value. Sample variance or s2 = SS/df SS: sum of squares df: degrees of freedom = n 1
How to compute for the t statistic
2. Estimate the standard error by substituting s2 in the formula for standard error. Note that SE is = M = /n or (2/n) Estimated standard error then is: SM = or SM =
How to compute for the t statistic
3. Compute the t statistic using the estimated standard error. t=
Comparing z and t formulas
The structure for the t formula is similar to that of the z-score: z or t = sample mean population mean (estimated) standard error For a hypothesis test, you hypothesize a value for the unknown population mean and plug that value into the equation along with the sample mean & estimated SE, which are computed from the sample data. If the hypothesized mean produces an extreme value for t, conclude that the hypothesis was wrong.
HYPOTHESIS TESTS WITH THE T STATISTIC
The problem Prof. X has prepared an Optimism Test that is administered yearly to graduating college seniors. The test measures how each graduating class feels about its futurethe higher the score, the more optimistic the class. Last years Batch Dungganan had a mean score of = 15. A sample of n = 9 seniors from Batch Harangdon was selected and tested. The sample mean was M=10 with SS=94. On the basis of this sample, can Prof. X conclude that Batch Harangdon has a different level of optimism than Batch Dungganan?
Steps in hypothesis tests with t-statistic
STEP 1: State your hypothesis. H0: = 15 (There is no change in the optimism level for this years graduating class.) H1: 15 (This years gradua ng class has a dierent optimism level than last years.)
Steps in hypothesis tests with t-statistic
STEP 2: Set your criteria. 2.a. Set alpha level & type of test. = .05, two-tailed t-test
Steps in hypothesis tests with t-statistic
STEP 3: Compute the test statistic. 3.a. Compute the sample variance. s2= or = = 11.75
2.b. Draw graph & locate the critical region using the ttable. With n = 9 students, df = n 1 = 8 For a 2-tailed test, with = .05, the critical t values are t = +/-2.306
STEP 3: Compute the test statistic. 3.b. Compute the estimated standard error using the sample variance. SM = = = 1.14
Steps in hypothesis tests with t-statistic
STEP 3: Compute the test statistic. 3.c. Compute the t statistic. t= = = 4.39
Steps in hypothesis tests with t-statistic
STEP 4. State your decision. t(8) = 4.39, p < .05 Thus, we reject H0. There is a significant difference in level of optimism between Batch Harangdon and Batch Dungganan.
EFFECT SIZE FOR THE T STATISTIC
Cohens d can be computed to measure effect size. The sample standard deviation is used in the formula to obtain an estimated value for d: estimated d = mean difference standard deviation =
Magnitude of d d = 0.2 d = 0.5 d = 0.8
Evaluation of Effect Size Small effect Medium effect Large effect
A second measure of effect size is r2, which measures the percentage of the variability that is accounted for by the treatment effect. r2 =
A directional test may be used in some research situations, such as exploratory investigations or pilot studies or when there is a priori justification (e.g., a theory or previous findings).
DIRECTIONAL & ONE-TAILED T TESTS
Seatwork Research on the effects of preschool childcare has found that children who spent time in day care, esp highquality day care, perform better on math and language tests than children who stay at home with their mothers. A sample of n=25 kids who attended day care before starting school had an average score of M=87 with SS=1536 on a standardized math test for which the population mean is =81. Is this sample sufficient to conclude that children with a history of preschool day care are significantly different from the general population? Use a (1) two-tailed test with =.05 (2) one-tailed test with =.05 Fig. 9-7, p. 298
