Evaluation activity

Evaluation activity

1. Why does a researcher use inferential statistics?

Inferential statistics are used to determine a more specific way of the probability of something (McMillan, Schumacher, 2010, pp316)

2. What effect does the existence of sampling error have on the analysis data?

sampling errors cause a variance in the data obtained as well as the means used to attain that data (McMillan, Schumacher, 2010, pp317)

3. What is the relationship between the null hypothesis and the research (alternative) hypothesis?

The null hypothesis has no difference in population means while the alternative hypothesis has a gap in population means. This means that the alternative theory states its research based on statistical terms unlike null hypothesis (McMillan, Schumacher, 2010, pp 317)

4. What does it mean to say the results of a study are statistically significant?

The results of a study are taken to be statistically significant when there is a rejection of the null hypothesis based on the alpha level ((McMillan, Schumacher, 2010, pp318)

5. What is the difference between the .10, .05, and .01 levels of significance?

If the value of p, which stands for probability is less than o.5, then the statement is said to be statistically significant. If it is between 0.5 and .10, it is said to be marginally substantial whereas if the figure is higher than .10 the value is said to be a non-significant difference (McMillan, Schumacher, 2010, pp320)

6. How can the results of a study be statistically significant yet practically non-significant?

This is when there is the use of a non-significant value and a significant value to get the conclusions of the experiment. Meaning that they combine .05 and .10 to get the results (McMillan, Schumacher, 2010, pp320)

7. What is the difference between a type 1 and type 11 error?

Type 1 error rejects a true null hypothesis while the type 11 error does not reject a false null hypothesis (McMillan, Schumacher, 2010, pp322)

8. How are independent samples different from dependent samples? What are the statistical consequences of using one or the other?

Dependent samples use one group for the researcher, while the independent samples have been involved in two groups. When using a dependent sample, one gets a true mean number at the end of the intervals, therefore, giving a more specific value. The independent samples give a statistical difference between the dependent variable and the two groups involved in the research (McMillan, Schumacher, 2010, pp323)

9. How does a t-test differ from an ANOVA?

A t-test involves a single group with two measures while analysis of variance (ANOVA) involves two or more groups ((McMillan, Schumacher, 2010, pp325)

10. How does a one-factor ANOVA differ from a factorial ANOVA?

A factorial ANOVA is an analysis of two or more independent variables placed together while a one-factor ANOVA involves the comparison of three or more variables on one independent variable (McMillan, Schumacher, 2010, pp325)

11. How does an ANCOVA differ from an ANOVA?

ANCOVA involves the adjustments for initial group differences while ANOVA involves two or more groups which give a more accurate probability statement (McMillan, Schumacher, 2010, pp326)

12. How does a MANOVA differ from ANOVA?

MANOVA has a two or more response variable unlike ANOVA which has one-way or two-way (McMillan, Schumacher, 2010, pp 327)

13. How do parametric samples differ from non-parametric statistics?

Parametric tests assume underlying statistical distributions in the data obtained while non-parametric tests do not necessarily rely on any distribution which means that they can be used even if parametric conditions are not yet reached (McMillan, Schumacher, 2010, pp328)

References

McMillan, J. H., & Schumacher, S. (2010). Research in Education: Evidence-Based Inquiry, MyEducationLab Series. Pearson.

References

McMillan, J. H., & Schumacher, S. (2010). Research in Education: Evidence-Based Inquiry, MyEducationLab Series. Pearson.