stats final study guide
A researcher wants to test if treatment adherence (Adhere/Not Adhere) is related to the number of chronic conditions (One condition/ More than one).
Contingency Table Chi-square test
Small expected frequency:
Contingency table chi-square also known as Chi-square of independence assume that expected frequencies should be at least 5 or greater. Smaller values of expected frequencies in each cell produces biased test result. The SPSS output provided a footnote in the chi-square table which help to identify this assumption.
Between-subjects designs:
Design in which different subjects serve under the different treatment levels. Examples: Independent sample t test and One-Way ANOVA are example of statistical test used in this type of design.
The chi-square test deals with:
Frequency data
A repeated-measures ANOVA differs from a paired sample t-test in that the paired sample t-test only deals with:
More than three times or conditions
Analysis of variance differs from t-test in that, ANOVA compares:
More than two means
The goodness-of-fit statistic is used when we have, how many categorical variables?
One
A regional manager was interested in examining the difference in the mean of sales profit of three stores: Store 1, Store 2, and Store 3. What statistical test would be the most appropriate to test differences between the three means?
One way-analysis of variance (ANOVA)
In the analysis of variance, if we get a significant F-statistic, we need to conduct which follow-up test to determine the differences between groups:
Post hoc test
. Multiple comparisons:
Techniques for making comparisons between two or more group means subsequent to an analysis of variance. Also known as post-hoc comparison or test. types: -Tukey HSD (honesty significant difference) test: -Fisher's least significance difference (LSD): -Bonferroni correction: -Games-Howell test:
Main effect:
The effect of one independent variable averaged across the levels of the other independent variable(s
Expected frequencies:
The expected values for the number of observations if the null hypothesis is true.
Observed frequencies:
The frequencies that were observed in the data collection
Odds:
The frequency of occurrence on one event divided by the frequency of occurrence of another event.
If research finds out that the effect of the type of program (online vs. Face-to-face) on performance depends on the instruction modality (synchronous vs. asynchronous), we can determine that:
The interaction effect between type of program and instruction modality is significant
Risk:
The number of occurrences on one event divided by the total number of occurrences of events
Familywise error rate:
The probability that a family of comparisons contains at least one Type I error
F statistic:
The ratio of MSgroup and MSerror. Test statistic for ANOVA
SS total:
The sum of squares of all the scores, regardless of the group of membership.
SS group (between):
The sum of squares of groups totals divided by the number of scores per group minus Σχ2 / N.
Sum of squares (SS):
The sum of the squared deviations (differences) around some point (usually a mean)
SS error:
The sum of the squared residuals.
MS within (MS error):
Variability among subjects in the same treatment group. Is the variability due to other factors.
MS between groups (MS group):
Variability among the group means. Is the variability due to the independent variable
a pediatrician is studying weight gain in infants. He divides them into 2 groups: breastfed and bottle-fed (Fed). Further, he divides them into those whose mothers feed them on a timed schedule, and those whoses mothers feed them when they cry (schedule). weight gain is the dependent measure. What type of analysis should you run?
a 2 x 2 factorial ANOVA
use the following research example to answer the next question: a researcher was interested in the effects of 1) alcohol consumption and 2) the content of a future expectations videotape, on the intention to quit school of male college students. The researcher randomly assigned 60 college-aged males to one of the following 3 groups (based on responses to a previous questionnaire): no alcohol consumed, a moderate amount of alcohol consumed, and a large amount of alcohol consumed. Additionally, half of the participants were shown an educational video about future expectations. the other half of the participants watched a documentary about owls (a control condition). At the end of the study, all the participants filled out a survey, all the participants filled out a survey related to intentions to drop out of school. higher scores on the survey indicated higher intentions to leave school. What type of statistical analysis would be most appropriate for this experiment?
a factorial analysis of variance
a goodness-of-fit chi-square test is used with:
a single categorical variable
picture 3
all of the above -a main effect of gender -a main effect of group - an interaction of gender x group
a repeated-measures analysis of variance differs forma one-way and a factorial design becuase:
all the above -the measures in a repeated-measures design are correlated -the meausres in a one way and factorial are independent -the meausres in a repeated-measures design are not independent from time to time
The major disadvantage of repeated-measures designs is that they:
are subject to the influence of carry-over effects
when we reject the null hypothesis in the analysis of varience we can conclude that
at least one of the means is different from at least one other mean
the chi-square test is used when we have:
categorical variables
picture 6 and 7
condition 1 and 4
a type of chi-square test in which we have two categorical variables and is used to test if both variables are associated is:
contingency table chi-square test
a researcher was interested in examining if wearing a face mask (yes/no) is contingent or conditional to the persons' age (young/adult). To test if these variables are independent of one another, what test should the researcher use to analyze the data?
contingency-table chi-square test
a researhcer wanted to test how movies influences subjects' IQ scores. she gave an IQ test to subjects following watching two movies. Half of the subjects first saw the Titanic followed by Schindler's List, while the other half first saw Schindler's list and then titanic. varying movie order is an example of
counterbalancing
the major difference between t tests and the analysis of varience is that the latter:
deals with mutiple groups (2 or more groups)
counterbalancing is a technique to:
distribute carry-over effects evely accross the data
the bonferroni procedure controls error rates by:
dividing the significance level (a) by the number of tests
When comparing differences in an experiment with two or more independent variables, which of the following test is more suitable?
factorial ANOVA
when comparing differences in an experiment with two or more independent variables we should use a(an):
factorial design
measuring the height of the same group of children every year for three years is an example of a between-subjects design
false
picture 8
false
the main difference in GPA based on gender and year in school is an example of the main effect
false
the mean difference in GPA based on gender is an example of an interaction effect
false
picture 9
goodness-of-fit chi-sqaure test
picture 4
graph B
picture 2
groups 1 and 2, groups 2 and 3, and groups 3 and 4
a repeated-measures design is ___ than the correspondin between-subjects design
more powerful
a factorial analysis of variance has:
more than one independent variable
an important assumption in the one-way analysis of variance is that:
observations are independent
we want to compare the scores of sifferent groups on a measure of reaction time. Three different groups were studied: patients with recent head injuries, patients with old head injuries, and a control group of non-injured peopole. we want to know which group of people has the fastest reaction time. what is the best statistical test to use to find this out?
one-way ANOVA
a researcher tested the effect of 3 different aversive conditions on fear. a group of 15 individuals participated in the 3 different conditions after which their fear was measure. the repeated measure ANOVA result was: F(2, 13)= 4.42, p=.034. what can we conclude?
participants' fear differs significantly over the three conditions
to interpret an interaction effect, we must:
plot the data in such a way that we see how eachh independent variable changes at each level of the other independent variable
What type of design represents a study in which we collected data from individuals who participated in a parenting program at three moments in time (Pre, Post, Follow-up)?
repeated measures
a ___ design is one in which subjects are measured repeatedly over time
repeated-measures
you want to run a study to examine the effects of poverty ont he development of antisocial behavior. You randomly select a large group of 12-year-old children and sort them into three gorups on the basis of family income. you meet them ever two years until the reach 20 years old, and measure antisocial behavior using a standard assessment. what test should you run to analyze this data?
repreated-measures ANOVA
the MS error in a repeated-measures design is ___ than the corresponding MS error in a between-subjects design
smaller
the main effect is defined as:
the effect of one independent variable averages across the level of the other independent variable
you should be careful about using a chi-square test when:
the expected frequencies are quite small (less than 5)
in a chi-square test the expected frequency is:
the frequency you would expect if the null hypothesis were true
use the following research example to answer the next question: a researcher was interested in the effects of 1) alcohol consumption and 2) the content of a future expectations videotape, on the intention to quit school of male college students. The researcher randomly assigned 60 college-aged males to one of the following 3 groups (based on responses to a previous questionnaire): no alcohol consumed, a moderate amount of alcohol consumed, and a large amount of alcohol consumed. Additionally, half of the participants were shown an educational video about future expectations. the other half of the participants watched a documentary about owls (a control condition). At the end of the study, all the participants filled out a survey, all the participants filled out a survey related to intentions to drop out of school. higher scores on the survey indicated higher intentions to leave school. the results indicated that the participants who watched the educational video scored significantly lower on the intention to drop out of school compared to the group that watched the owl video. what does this suggest?
the is a significant main effect of videotape
in the analysis of variance we will assume that:
the populations are normally distributed
the familywise erroe rate is:
the proability of at least one type 1 error
In the analysis of varience with three groups the null hypothesis is:
the three population means are equal to each other
picture 1
there are statistical differences among the groups, F(4,392)= 170.897, p<.05
picture 5
there is a significant difference accross conditions with a large effect size
use the following research example to answer the next question: a researcher was interested in the effects of 1) alcohol consumption and 2) the content of a future expectations videotape, on the intention to quit school of male college students. The researcher randomly assigned 60 college-aged males to one of the following 3 groups (based on responses to a previous questionnaire): no alcohol consumed, a moderate amount of alcohol consumed, and a large amount of alcohol consumed. Additionally, half of the participants were shown an educational video about future expectations. the other half of the participants watched a documentary about owls (a control condition). At the end of the study, all the participants filled out a survey, all the participants filled out a survey related to intentions to drop out of school. higher scores on the survey indicated higher intentions to leave school. What would you conclude if the researcher found that alcohol consumption increased intention to drop out of school, but only when the participants had watched the owl video?
there is a significant interaction between alcohol consumption and videotape
the major advantage of repeated-measures designs is that:
they allow you to remove subject differences from the error term
in multiple comparison procedures, post-hoc tests are completed after the ANOVA. Why are post-hoc tests preferred over runnning several t-tests?
they decrease the probability of a type 1 error
F is the ratio of MS group divided by MS error
true
a chi-square statistic test compares the observed frequencies with the expected frequencies. T/F
true
a within-subject design is when the same pariticant serce in the same condition
true
an ANOVA is used to compare the means of two or more groups
true
in an analysis of variance, factor refers to independent variables
true
the odds refer to the frequency of occurrence of one event divided by the frequency of occurrence of another event. T/F
true
Within-subjects designs: also called:
-Experimental designs in which each participant produces multiple scores. -repeated-measures designs
in a repeated-measures ANOVA, the test to correct the degrees of freedom, such as Greenhouse-Geisser and Huynh-feldt, should be used if:
... you do not need to correct the degrees of freedom
if we run six independent comparisons among means, each at the 5% level, the overall familywise error rate will be approximately
.30
a teacher found significant differences in the mean running speeds of sprinters wearing shoes made by nike, reebok, and adidas using an analysis of varience. the n^2 caluclated on the basis group membership (based on which shoes are worn) equaled .16. the value of n^2 shows that:
16% of the variability in running speed is attributed to shoe brand
a 2 x 4 factorial has:
2 levels of one variable and 4 levels on the other
Omega squared (ω2 ):
A less biased measure of the magnitude of effect compare to the etasquared.
Eta-squared (η 2 ):
A measure of the magnitude of the effect typically used in ANOVA. Eta squared is the ratio between the variability among observations that can be attributed to group effects (SSgroup) and the total variability (SStotal). Eta squared is the ratio between the variability among observations that can be attributed to group effects (SSgroup) and the total variability (SStotal).
Tukey HSD (honesty significant difference) test:
A multiple comparison procedure for making pairwise comparisons among means while holding the familywise error rate at α=.05.
Bonferroni correction:
A multiple comparison procedure in which the familywise error rate is divided by the number of comparisons.
Fisher's least significance difference (LSD):
A multiple comparison technique that requires a significant overall F, and that involves standard t test between pairs of means.
Games-Howell test:
A multiple comparison technique used when the assumption of homogeneity of variance is not met
Interaction effect:
A situation in factorial design in which the effect of one independent variable depends on the level of another independent variable.
Analysis of variance
A statistical technique for testing for differences in the means of several groups (two or more)
Chi-square test:
A statistical test often used for analyzing categorical data.
Goodness-of-fit test:
A test of comparing observed frequencies with expected frequencies.
Fisher's Exact Test:
A test on a contingency table that assumes fixed marginal means.
Contingency table:
A two-dimensional table in which each observation is classified based on two variables simultaneously.
One-way ANOVA:
An analysis of variance where the groups are defined on only one independent variable. Analyze differences in two or MORE independent means.
Counterbalancing:
An arrangement of treatment conditions designed to balance out practice effect.
repeated-measures designs:
An experimental design in which each subject receives all levels of at least one independent variable.
Two-way factorial design (2X2):
An experimental design involving two independent variables in which every level of one variable is paired with every level of the other variable. Also called factorial design.
the difference between a one-way analysis of variance and a factorial analysis of variance is
Both a and b -the presence of an interaction -the presence of more than one main effect