T-test ~ Sample Problems
Case 1 Comparing Sample Mean with Population Mean
Example Problem: Suppose we want to test, on the basis of a random sample of size n=5, whether
the fat content of a certain kind of processed meat exceeds 30%. What can we conclude at the 0.01 level of significance, if the sample values are 31.9, 30.3, 32.1, 31.7, and 30.9%? Step 1: The fat content of a certain kind of processed meat is less than or equal to 30%. The fat content of a certain kind of processed meat exceeds 30%. Step 2: = 0.01 Step 3: Since n < 30, the appropriate test statistic is t-test with df = n-1 = 5-1 = 4 Step 4: df = 4 ; = 0.01 Critical Values of t df 1 4 0.05 0.025 0.01 0.005
3.74
Step 5:
(
Step 6: Since |CV| ? |TV|
is greater than |4.08| ? |3.74|
, the null hypothesis must be rejected and the
4.08 > 3.74 .: Reject and Accept
alternative hypothesis must be accepted. In other words, the mean fat content of the given kind of processed meat exceeds 30%.
SOURCE: Title of the book: STATISTICS A First Course Author: John E. Freund et. al.
Case 2 Comparing Two Independent Sample Means
Example Problem: I five games with relatively light ball, a professional bowler scored 205, 220,
200, 210, and 201, and in five games with a somewhat heavier ball he scored 218, 204, 223, 198, and 211. At the 0.05 level of significance, can we conclude that on the average he will score higher with the heavier ball? Step 1: is the average score scored with the light ball. is the average score scored with the heavier ball. The professional bowler will score lower or just the same with the heavier ball. The professional bowler will score higher with the heavier ball. Step 2: = 0.05 Step 3: Since n < 30, the appropriate test statistic is t-test with df = Step 4: df = 8 ; = 0.05 Critical Values of t df 1 8 0.05 0.025 0.01 0.005 = 5+5-2 = 8
Step 5:
Step 6: Since is less than
( )
, the null hypothesis must be accepted and the
alternative hypothesis must be rejected. In other words, the professional bowler will score lower or just the same with the heavier ball. |CV| ? |TV| |-0.62| ? |-1.860| 0.62 < 1.860 .: Accept and Reject
SOURCE: Title of the book: STATISTICS A First Course Author: John E. Freund et. al. .
Case 3: Comparing Two Dependent or Correlated Sample Means
Example Problem: Suppose before and after two weeks of a reading and typing course as a part of
the OJT, we record the scores of five people in 20-minute typing test. Before 214 467 678 511 579 After 245 534 712 529 609
Ben Ted Bob Sarah Betsy
their students? Step 1:
With 0.05 as the level of significance, can we prove that the course improves the typing speed of
The course does not improve the typing speed of their students. The course improves the typing speed of their students. Step 2: = 0.05 Step 3: Since n < 30, the appropriate test statistic is t-test with df = n-1 = 5-1 = 4 Step 4: df = 4 ; = 0.05 Critical Values of t df 1 4 0.05 0.025 0.01 0.005
Step 5: Before 214 467 678 511 579 Total After 245 534 712 529 609 d 31 67 34 18 30 180
Ben Ted Bob Sarah Betsy
961 4489 1156 324 900 7830
( ) )
( (
( )
Step 6: Since their students. |CV| ? |TV| |4.38| ? |2.132| 4.38 > 2.132 .: Reject and Accept is greater than
, the null hypothesis must be rejected and the
alternative hypothesis must be accepted. In other words, the course improved the typing speed of
SOURCE: Title of the book: Basic Business Statistics Author: Daniel Sankowsky
