Exam 3 Flashcards
What does it mean to test for differences?
difference testing or analysis is used to see where there are differences between or among groups of people or texts
What does it mean that groups are significantly different on a measure or variable?
this means that the differentiations noted in the results of the data had a cause and did not occur by chance
What does the null hypothesis say about differences?
in order for there to be a statistically significant difference, the statistical difference has to be less than 0.05 (p =< 0.05)
Type of Nominal question
select which assessment practice you prefer for providing feedback (written, digital, or audio)?
Type of Ordinal question
rank order your favorite to least favorite method for providing feedback (written, digital, audio)
Type of Interval question
On a scale of 1-5, how likely are you to write out your feedback by hand?
Type of Ratio question
For how many assignments, do you provide typed out feedback?
What test is used for nominal questions?
chi-square test
What test is used for ordinal questions?
median test
What test is used for interval questions?
t-test and analysis of variance (ANOVA)
What test is used for ration questions?
t-test and analysis of variance (ANOVA)
What are 3 Requirements for Statistical Significance?
significance level, degrees of freedom, and critical value
What does significance level mean?
the probability for rejecting the null hypothesis which can be equal to or less than 0.05 (p =< 0.05)
What does degrees of freedom mean?
the number of scores that are free to vary?
How is the degree of freedom calculated?
it varies depending on the type of test
What does critical value mean?
the number your test results must meet or beat for them to be considered significant
How is the critical value calculated?
use the degrees of freedom # and significance level to look it up using Harris (1998) chart
Example of 1-tailed RQ/hypothesis that would require use of difference testing
Is there a difference in levels of confidence providing students with feedback between teachers with 2 years of teaching experience and teachers with 5 years of teaching experience? I predict that teachers with 5 years of experience will have more confidence in providing feedback than teachers with 2 years of experience.
Example of 2-tailed RQ/hypothesis that would require use of difference testing
Is there a difference in levels of confidence providing students with feedback between teachers with 2 years of teaching experience and teachers with 5 years of teaching experience? I predict that there will be differences between the levels of confidence amongst teachers with 5 years of experience and 1 year of experience.
In the following question, which is the IV (grouping value) and which is the DV (variable measured):
Is there a difference in levels of confidence providing students with feedback between teachers with 2 years of teaching experience and teachers with 5 years of teaching experience?
The IV is 2 groups: teachers with 5 years of experience and teachers with 2 years of experience.
The DV is levels of confidence.
For what type of data can a Chi-square be used for?
nominal
What is the difference between a single sample Chi-Square test and a multiple sample Chi-Square test?
The one-variable chi-square test is used when there is 1 IV or 1 DV being tested.
The two-variable chi-square test is used when there are 2 variables being tested such as 2 IVs or 1 IV and 1DV.
What are the 4 steps for a one-variable chi-square test
*List the observed frequency (O) for each category
*List the expected frequency (E) for each category (to calculate this, divide the # of observations by the total # of categories
*For each category:
-subtract the expected from the observed (O-E).
-Square the difference.
-Divide the difference with the expected frequency (E)
*Add up the results for each category to get the chi-square value
How is the degree of freedom calculated in a chi-square?
subtract 1 from the number of categories (n-1)
If the chi-square value is 22.4, the degrees of freedom is 3, and the critical value is 7.815, is there a significant value? And how would you write it out in an article?
Because the chi-square value is 22.4, which is greater than the critical value of 7.815, the null hypothesis can be rejected. Hence, the study’s hypothesis that there is a significant difference between the variables can be accepted.
What is an example of a RQ/Hypothesis that would require a Chi-Square test?
What is the difference in distribution among the assessment practice teachers prefer for providing feedback (written, digital, or audio)? I predict that teachers preference will vary with most picking digital feedback.
What is a Median test?
A test used to test differences among ordinal data
What is an example of a RQ/ hypothesis that would require a Median test?
Is there a statistical difference between which feedback method (written, digital, audio) that teachers perceive to be the most effective? Teachers will perceive digital to be a more effective feedback method
What are the two types of t-tests and for what type of data is it used?
Paired t-test and independent sample t-test. It is used for interval and ratio data
What is a paired t-test and an example?
a test used to compare results of a pretest and post test. For example, what is the height difference of a person at age 5 versus the height of a person at age 15
What is an independent sample t-test and an example?
a test used to compare 2 groups. For example, what is the height difference between 5 males and 5 females
What is an example of a RQ/Hypothesis that would require a t-test?
Is there a difference between teachers with 5 years of experience and 1 year of experience in terms of their calculated level of confidence in assessing writing? I predict that teachers with 5 years of experience will perceive their abilities to assess student writing with higher confidence.
What are the 4 steps for calculating an independent t-test?
-Calculate the mean for each group.
-Calculate the standard deviation for each group
-Calculate the degree of freedom
-Apply to the formula
If the t-test results are 2.5, the degree of freedom is 8, p <= .05, and the critical value is 2.33, is there a significance difference?
Results indicate that X is better than Y because the t-test results are higher than the critical value (t= 2.5, df= 8, p<= .05)
What is the difference between the grouping variable (IV) and the dependent variable in an ANOVA study?
the IV represents the way people/ texts are grouped/ classified (e.g. group 1, 2, 3). The DV represents the variable that is being tested to see if there is a difference among the groups (e.g. level of confidence)
How many IVs and how many DVs for a One-Way Analysis of variance?
there are 1 IVs and 1 DV
What is an example of a RQ/Hypothesis that would require a One-Way ANOVA?
Are there differences in student performance when a teacher uses written, audio, or digital feedback? Teachers who provide written feedback will have students who perform higher than teachers who provide audio or digital feedback.
What does Between Group Variance in the context of an ANOVA mean?
this means that there are differences between the IV groupings in relation to the DV being tested
What does Within Group Variance in the context of an ANOVA mean?
This means that within a single IV grouping there are differences among the participants in terms of relation to the DV being tested. The results could be skewed and far apart or close and similar.
When do ANOVA results mean that the hypothesis can be accepted?
when the calculated value exceeds the critical value
What numbers do you report when writing up One-Way ANOVA results?
(F = #, df = #, p<= #). F is the ANOVA result, df is the degree of freedom, and p is the significance level
What are 3 post-hoc analyses and why do you need to use them with One-way ANOVA?
*Scheffe ⇒ conservative = this affects the ANOVA results by making changes to the significance level to protect from making a Type 1 error.
*Tukey ⇒ moderate = this affects the ANOVA results by not making much changes to the significance level.
*LSD ⇒ liberal = this affects the ANOVA results by having a lot of leeway with the significance level implying there is a more likely chance of making a Type 1 error.
If the ANOVA results are 4.78, the degree of freedom is 2 (numerator) and 12, p <= .05, and the critical value is 3.885, is there a significance difference?
Results indicate that there is a difference between X, Y, and Z in terms of [topic] because the ANOVA results are higher than the critical value (F= 4.78, df= (2, 12), p<= .05)
What is a Factorial ANOVA and when do you use it?
You use it when there are 2 IV that need to be tested.
For example, is there a difference in strength of 2 rebuttal strategies?
Is there a difference in the strategy when used in 4 different topics?
How many IVs and how many DVs for a Factorial ANOVA?
There are 2 IVs and 1 DV
What is a Multivariate Analysis of Variance (MANOVA) and when do you use it?
you use it when you have 2 or more DVs
