One Population Test

Download as pdf or txt
Download as pdf or txt
You are on page 1of 15

One Population Test

Dr. Elmanani Simamora, M.Si

DR. ELMANANI SIMAMORA, M.SI 1


Your best quote that reflects your
approach… “It’s one small step for
man, one giant leap for mankind.”

- NEIL ARMSTRONG

DR. ELMANANI SIMAMORA, M.SI 2


One Population Test
Purpose of the Practicum
Test the average difference of a group with a hypothesis or the similarity of the
average of a group of data with a certain average with One Sample t Test.

Educational Research Issues


A teacher wants to compare the math scores of all grade VII students with the
average of last year's exam scores. The average exam last year was 6.5. Is the
average value significantly different from the 6.5 average or not?

DR. ELMANANI SIMAMORA, M.SI 3


One Population Test
When do we use the t-test?
We use the t-test if the following conditions:
* Sample size is relatively small (sample size is less than thirty)
* Population variance is unknown

When do we use the Z-test?


We use the t-test if the following conditions:
* Sample size is relatively large (sample size is more than thirty)
* Population variance is known (available)

DR. ELMANANI SIMAMORA, M.SI 4


One Population Test
The following are data presented on the National Examination
(UN) scores for grade VIII mathematics subjects (The Table
beside).

Is the average score for this year's Mathematics UN


significantly different from last year's Mathematics UN
average? It is known that the average of mathematics last year
was 69. (Use a significance level of 5%)

DR. ELMANANI SIMAMORA, M.SI 5


One Population Test
SPSS Procedure
1. Run the SPSS 22 program, select Variable View at the bottom.
2. Fill in the column name "UN Math" in the first row with decimals worth 2
and interval data scale (select from fiture of measure “Scale”).

DR. ELMANANI SIMAMORA, M.SI 6


One Population Test
SPSS Procedure
3. Select Data View next to Variable View and enter data UN scores in
Mathematics.

DR. ELMANANI SIMAMORA, M.SI 7


One Population Test
SPSS Procedure
4. Select from the following Analyze menu
Analyze → Compare Mean → One Sample T Test
5. Enter the UN Math variable into the Test Variable (s) box and enter 69 in
the Test Value box.

DR. ELMANANI SIMAMORA, M.SI 8


One Population Test
SPSS Procedure

DR. ELMANANI SIMAMORA, M.SI 9


One Population Test
SPSS Procedure
6. Click the Option button then the One-Sample T Test: Option dialog box
appears. By default the Confidence Interval is 95% and Missing Values is
selected Exclude case analysis by analysis which means only valid data is
used in the analysis.
7. Click the Continue button.
8. Click OK

DR. ELMANANI SIMAMORA, M.SI 10


One Population Test
SPSS Procedure

DR. ELMANANI SIMAMORA, M.SI 11


One Population Test
Output of SPSS

DR. ELMANANI SIMAMORA, M.SI 12


Readings of Analysis Results
• The One-Sample Statistics table displays descriptive statistical analysis results
that contain valid data (N), average values, standard deviations, and average
standard errors.
• The One-Sample Test table displays the results of the analysis of the test of
the average difference of a group with a hypothesis.
Research Hypothesis:
Ho: The average for this year's UN Math is the same as last year.
H1: The average of this year's UN Math is not the same as last year.
Equal to
Ho: 𝜇 = 69
H1: 𝜇 ≠ 69

DR. ELMANANI SIMAMORA, M.SI 13


Readings of Analysis Results
• Criteria of Statistical Tests
If t-count <t-table, then Ho is accepted. If t-count> t-table, then Ho is
rejected.
Or (equal to)
if Sig (2-tailed)> ½ α, then Ho is accepted. If Sig (2-tailed) <½ α, then Ho is
rejected.
• Conclusion
Sig (2-tailed = 0.100)> ½ α (0.025), then Ho is accepted. So the average
Mathematics UN for this year is the same as last year or there is no significant
difference in the average score of this year's Natural Sciences UN and last
year.

DR. ELMANANI SIMAMORA, M.SI 14


The Task of Blended Learning’s Practicum
Make one problem / case in the world of education that is equipped with data
(fictitious data does not matter) and do a comparative test with One Sample T
Test (use a significance level of 5% with respondent data totaling at least 30) and
Z-score.

Send your answers to emails until 6:30 PM on April 6, 2020 by Classroom

DR. ELMANANI SIMAMORA, M.SI 15

You might also like