Scribd
Upload
13 views9 pages

(RM1) Key-Statistical Problems

The document presents various statistical analyses including regression, independent sample t-tests, ANOVA, and correlation tests to evaluate predictors of job performance, vocabulary scores, instructional methods, and recall performance. Key findings include that job aptitude is not a significant predictor of performance, while there is a significant difference in vocabulary scores between experimental and control groups. Additionally, the effectiveness of instructional methods varies between psychologists and psychiatrists, and there are no significant differences in recall performance between rereading and paraphrasing.

Uploaded by

KTMD
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
13 views9 pages

(RM1) Key-Statistical Problems

The document presents various statistical analyses including regression, independent sample t-tests, ANOVA, and correlation tests to evaluate predictors of job performance, vocabulary scores, instructional methods, and recall performance. Key findings include that job aptitude is not a significant predictor of performance, while there is a significant difference in vocabulary scores between experimental and control groups. Additionally, the effectiveness of instructional methods varies between psychologists and psychiatrists, and there are no significant differences in recall performance between rereading and paraphrasing.

Uploaded by

KTMD
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 9

Answer Q1: Regression (Linear)

- Whether job aptitude is a good predictor of performance:


p = Sig. = 0.103 (>0.05) → accept null hypothesis → There is no statistical significance → Not a
good predictor

- How much variation in the job performance is due to job aptitude: 29.7%
R² = 0.297 → This means that 29.7% of the variation in Job Performance is explained by Job Aptitude.

- The regression equation: y = a + bx (DV = Intercept + Slope x IV) → y = 48.890 + 0.494 x X


- Intercept: 48.890
- The slope: 0.494

- Job aptitude = 70 → Performance test = 48.890 + 0.494 x 70 = 83.470

Answer Q2: INDEPENDENT SAMPLE T-TEST


● The Experimental group has a higher mean vocabulary score (223.0833) compared to the Control group
(201.0833).
● The standard deviation suggests more variability in the Experimental group.

● p = 0.258. Since p > 0.05, we fail to reject the null hypothesis, meaning we assume equal variances for the
two groups.

This means we will use the first row ("Equal variances assumed") for interpretation.

● t(22) = 2.266, p = 0.034

Since p < 0.05, we reject the null hypothesis, meaning there is a statistically significant difference between
the experimental and control groups.

● Mean Difference = 22.00 → The experimental group had a vocabulary 22 words higher on average
than the control group.
● Standard Error of Difference = 9.71 → The variability in the difference between means.
● 95% Confidence Interval = [1.87, 42.13] → Since the confidence interval does not include zero, this
confirms that the difference is statistically significant.

1.00 1.00 75.00


1.00 1.00 94.00
1.00 1.00 81.00
1.00 2.00 88.00
1.00 2.00 93.00
1.00 2.00 63.00
1.00 3.00 96.00
1.00 3.00 94.00
1.00 3.00 97.00
2.00 1.00 74.00
2.00 1.00 93.00
2.00 1.00 82.00
2.00 2.00 82.00
2.00 2.00 95.00
2.00 2.00 99.00
2.00 3.00 74.00
2.00 3.00 63.00
2.00 3.00 79.00
To determine if the three methods of instruction (Lecture, Television, Internet) significantly differ in effectiveness for
Psychologists and Psychiatrists, we will perform a Two-Way ANOVA with:

● Independent Variables (IVs):


○ Method of Instruction (Lecture, Television, Internet)
○ Profession (Psychologists, Psychiatrists)
● Dependent Variable (DV):
○ Knowledge Test Scores

● The Sig. (p-value) for "Based on Mean" = 0.124


● The Sig. (p-value) for "Based on Median" = 0.799
● The Sig. (p-value) for "Based on Trimmed Mean" = 0.153

Decision Rule:

● The null hypothesis (H₀) for Levene’s test states that the variances are equal across groups.
● If p > 0.05, we fail to reject H₀, meaning the assumption of homogeneity of variance is met.

Meth (Method of Instruction):

● F(2,12)=0.210, p = .813 → Not significant.


● Interpretation: The method of instruction (Lecture, TV, Internet) does not significantly affect the test scores.

Job (Psychologists vs. Psychiatrists):

● F(1,12)=0.901F(1,12) = 0.901F(1,12)=0.901, p = .361 → Not significant.


● Interpretation: There is no significant difference between Psychologists and Psychiatrists.

Meth * Job (Method × Job Interaction):

● F(2,12)=4.673F(2,12) = 4.673F(2,12)=4.673, p = .032 → Significant at α = .05.


● Interpretation: There is a significant interaction effect, meaning the effectiveness of instruction methods differs
between Psychologists and Psychiatrists.
● Effect Size (Partial Eta Squared = .438): This suggests a large effect size (above .14 is considered large).
→ Even though neither Method nor Job alone significantly affects scores, their combination does.

→ This means that one method might work better for Psychologists while another works better for
Psychiatrists.

● All p-values (Sig.) are greater than .05, meaning none of the method comparisons are statistically
significant.
● Example: The difference between Method 1 and Method 2 is not significant (p=.817p = .817p=.817).
● Even though your ANOVA interaction effect (Meth * Job) was significant (p=.032p = .032p=.032), your
Tukey post-hoc test does not find significant pairwise differences between individual methods.
● This suggests that while the interaction effect exists, the specific group differences are not large enough
to be individually significant.

Answer Q4: z-test


Answer Q5: Independent sample t-test
● p-value of Levene’s test is 0.818 > 0.05 (use the first row) → equal assumed variances
● p-value (of the 1st row) is 0.101 > 0.05 → accept hypothesis → there is NO statistical significance
→ This means there is no significant difference in test scores between students who studied with
background music and those who studied without it.

Answer Q6: paired t-test

● The negative mean difference (-1.30) suggests that paraphrasing led to slightly higher recall scores
compared to rereading. → This means that, on average, students scored 1.30 points lower when
rereading compared to paraphrasing.
● However, the confidence interval includes zero (-3.215 to 0.615), indicating the true difference could be
positive, negative, or zero.
● The t-value (-1.421) is not large enough to show a significant effect.
● The p-value (0.172) is greater than 0.05, meaning we fail to reject the null hypothesis.
→ This means there is no significant difference between rereading and paraphrasing in recall performance.

Answer Q7: Correlation (Analyze → Correlate → Bivariate)


● The Pearson correlation coefficient (r) between English and Filipino grades is 0.718 (close to +1). → This
indicates a strong positive correlation, meaning students who perform well in English tend to perform well in
Filipino.
● The p-value = 0.019.
● Since 0.019 < 0.05 (α level), the correlation is statistically significant. → This means there is a low
probability (1.9%) that the observed relationship is due to chance.
● Since r = 0.718 is greater than the critical value r = 0.632, and p = 0.019 is less than 0.05, we reject the null
hypothesis.

Answer Q8: Linear regression

● Equation of the least square line: y = 41.873 + 0.39 x X


● Extent and direction of the relationship: coefficient r = 0.324 < 1 → weak positive relationship
between abstract reasoning and verbal ability
● Value of the coefficient of the determination: R2=0.105 → only 10.5% of the variation in verbal ability is
explained by abstract reasoning.
● How much of the variation in verbal ability can be explained by factors other than the abstract reasoning: The
remaining 89.5% (100% - 10.5%) of the variation in verbal ability is due to factors other than abstract
reasoning.
● x = 46 → y = 41.873+0.390 x 46 = 59.813
● p-value for the regression (Sig. in ANOVA) = 0.239 (greater than 0.05) → The model is not statistically
significant, implying that abstract reasoning is not a strong predictor of verbal ability.

Answer Q9: One-way ANOVA. One-way ANOVA since there is one DV (scores) and one ID (section/teachers)

● The p-value (Sig.) for Levene's test is 0.244 (based on mean). Since 0.244 > 0.05, we fail to reject the null
hypothesis of equal variances.
● p-value (0.011) is less than 0.05 → reject the null hypothesis that all group means are equal.
Conclusion: There is a statistically significant difference in scores between at least one of the groups.
Answer Q10: Two-way ANOVA (Analyze → General Linear Model → Univariate). 2x2 factorial ANOVA

● Sig. = 0.866 (Based on Mean) → fail to reject the null hypothesis, meaning that the assumption of
homogeneity of variances is met. This means we can trust the ANOVA results.

● The size of the lecture room alone does not significantly affect test scores (p = 0.357).
● The size of the test room alone does not significantly affect scores (p = 0.721).
● There is a strong interaction between lecture room and test room (p<.001).. This means the effect of the
lecture room depends on the test room size.

You might also like