Deck 12: Confidence Intervals and Effect Size: Building a Better Mousetrap

Question 1

A single number that is used to predict another value is a point estimate.

Answer 1

True

Question 2

Confidence intervals that have higher % confidence are likely to encompass a wider range than one that has a lower % confidence.

Answer 2

True

Question 3

Confidence Intervals are created for the sole purpose of estimating the true population mean.

Answer 3

False

Question 4

Precision is best described as when the interval estimate captures the true estimated value.

Question 5

The null and alternate hypotheses generally are about different distributions.

Question 6

If an IV has a significant effect on a DV, then the effect size must be large

Question 7

Cohen's d is an estimate of the true population mean.

Question 8

Cohen's d can be used to estimate the size of a mean difference between two groups

Question 9

when using a two-way ANOVA, there are exactly 3 measurable partial -eta-squared.

Question 10

With respect to the value of Cohen's d, effects that are small are near 0, and effects that are considered large must be greater than 1.

Question 11

A t-test found a significant effect in which one condition had a detrimental effect on the outcome, making it significantly worse. In this case, one would expect that Cohen's d would be negative.

Question 12

Eta-squared cannot be greater than 1.00.

Question 13

eta-squared cannot be less than 0

Question 14

Partial eta-squared are only used for factorial ANOVAs, and not one-way ANOVAs.

Question 15

Like an inferential test, confidence intervals can be calculated on mean differences to test for statistical significance..

Question 16

When the 95% confidence interval of the mean of one group does not include 0, it means that the difference between two groups is statistically significant at the .05 level

Question 17

Effect sizes and confidence intervals becoming less and less frequently used in published research. They are more common for oral presentations.

Question 18

One reason why effect sizes are generally omitted is because it makes the statistics harder to understand and does not provide any additional information that significance testing doesn't already tell us.

Question 19

The 99% confidence interval of a mean difference is from .04 to 1.49. This suggests that the mean difference is significant at the alpha = .05 level.

Question 20

As confidence increases, precision generally decreases

Question 21

When one is certain that an interval is a good one, one likely has high ______________.

A) confidence
B) validity
C) point estimate
D) none of the above

Question 22

How do point estimates compare to confidence intervals?

A) Correct point estimates would signal high precision, but point estimates tend to have lower confidence levels than do confidence intervals
B) Point estimates are more accurate than confidence intervals.
C) Point estimates are better for interval and ratio scales, where as confidence intervals are only useful for interval scales.
D) All of the above are correct ways to compare the two.

Question 23

When calculating a confidence interval for a independent samples t-test instead of a one-sample t-test, which of the following best describes the difference in their calculations?

A) Instead of the population standard deviation, you use the sample standard deviation.
B) Instead of the population mean, you use the sample mean.
C) You use difference scores instead of the sample mean.
D) The critical values are different

Question 24

In the context of ANOVA, an eta-squared is used to portray which of the following?

A) The degree of statistical significance of mean differences
B) The necessary sample size to obtain a statistically significant result
C) The proportion of error variability in the population the study failed to account for.
D) The proportion of variability explained by one variable on another variable.

Question 25

Use the data below for the following problem.
$\begin{array}{lcc}\hline \text { Source } & S S & d f& MS &F \\\hline \text { SSeffect } & 244.5 & 2 \\\text { SSerror } & 1436.7 & 23\\\hline\end{array}$ What is the eta-squared?

A) .16
B) .15
C) .17
D) .09

Question 26

A doctor is examining differences between two treatments. He calculated a 95% confidence interval on the mean difference between the two groups. The interval ranged from -1.2 to 3.48. What does this suggest?

A) The mean difference is not significant at the .05 level
B) The mean difference has a small effect size
C) The mean difference is significant at the .05 level, but not at the .01 level.
D) The mean difference is not significant at the .05 level, but could still be significant at the .01 level.

Question 27

In general, when you calculate a 95% confidence interval instead of a 90% confidence interval, which of the following changes?

A) The width of the interval expands
B) The interval is more precise
C) The interval is more likely to capture the true value.
D) All of the above

Question 28

When calculating a Cohen's d in a dependent samples t-test what does it tell us?

A) The extent the two groups are correlated
B) The size of the difference in standard deviation units
C) The level of statistical significance
D) The probability the independent variable had a true effect on the dependent variable.

Question 29

What is the relationship between effect size and statistical significance?

A) Significant results have meaningful effect sizes. Effect sizes are meaningless without statistical significance.
B) Significance at alpha = .05 is always a smaller effect than significance at alpha = .01
C) There is no relationship between statistical significance and effect size whatsoever.
D) All else equal, larger effects are more likely to be significant.

Question 30

Point estimates can be used to estimate which of the following values?

A) Population standard deviation
B) Population mean
C) The true effect size of a mean difference
D) All of the above

Question 31

Which of the following constitutes a medium effect for a Cohen's d?

A) .3
B) .1
C) .9
D) .5

Question 32

Which of the following most accurately describes A 95% confidence interval of a mean?

A) the sample has a 95% probability of capturing the true mean.
B) 95% of all samples will capture the true mean
C) 95% of the time, the true mean will be in the sample statistic
D) 95% of the sample means will be statistically significant.

Question 33

A researcher is interested whether an intervention successfully decreases prejudice in majority populations. When conducting an independent samples t-test, a researcher calculates the t-statistic with an alpha level of .05, and a 95% confidence interval of the mean difference. Which of the following is most likely true?

A) both procedures will lead to the same conclusion regarding statistical significance
B) both procedures will lead to a point estimate
C) both procedures may lead to difference results regarding whether the difference is statistically significant. This is because the traditional t-statistic is a point estimate and the confidence interval is an interval estimate.
D) The t-statistic and confidence interval will tell us fundamentally different information. The t-statistic compares mean differences, whereas the confidence interval will estimate the true effect size of the difference.

Question 34

In a 3 x 4 ANOVA, how many partial eta-squared could one calculate?

A) 4
B) 3
C) 12
D) 2

Question 35

$\begin{array}{lcc}\hline \text { Source } & S S & d f& MS &F \\\hline \text { Factor 1 } & 44325 & 3 \\\text { Factor } 2 & 32241 & 2 \\\text { Factor }{ }^{*} \text { Factor 2 } & 9382 & 6 \\\text { Error } & 76257 & 24\\\hline\end{array}$ How is this ANOVA best characterized and how many partial eta-squared are there for this factorial ANOVA?

A) 3 x 2 ANOVA with 3 partial eta-squared
B) 3 x 2 ANOVA with 2 partial eta-squared
C) 4 x 3 ANOVA with 3 partial eta-squared.
D) 2 x 2 ANOVA with 3 partial eta-squared

Question 36

$\begin{array}{lcc}\hline \text { Source } & S S & d f& MS &F \\\hline \text { Factor 1 } & 44325 & 3 \\\text { Factor } 2 & 32241 & 2 \\\text { Factor }{ }^{*} \text { Factor 2 } & 9382 & 6 \\\text { Error } & 76257 & 24\\\hline\end{array}$ What is the partial eta-squared for factor 1?

A) .58
B) .37
C) .38
D) .13

Question 37

A researcher collects a sample of individuals' productivity by measuring the number of words they typed in an hour during a writing session with 20 individuals. The sample mean is 1738.2 words this week. The individuals from last week's 20 individuals only had a mean of 1553.7 words. The pooled variance is 11545.9.
Calculate the Cohen's d for this mean difference.

A) .02
B) 1.7
C) insufficient information
D) -.02

Question 38

The researcher calculates a 95% confidence interval for the mean difference. 0 is not captured in the interval. What would this suggest?

A) The mean difference is not significant
B) The mean difference is significant
C) The mean difference has little to no effect
D) The data is skewed because there are no negative values in the confidence interval of the mean difference.

Question 39

If you construct a 95% confidence interval around a sample mean and the 95% CI does not include the population mean, what would this suggest?

A) the observed mean is significant larger than the population mean
B) the observed mean is significantly smaller than the population mean
C) the observed mean is significantly different from the population mean
D) the observed mean is not significantly different from the population mean.

Question 40

Cohen's d is the effect size often reported for an independent samples t-test. How does the calculation for Cohen's d compare to the calculation of a t-statistic?

A) Cohen's d uses pooled standard error and the t-statistic does not
B) Cohen's d is not directly influenced by the size of the sample, but the t-statistic is.
C) Cohen's d uses the pooled standard deviation whereas the t-statistic does not require a measure of variability.
D) Cohen's d is a measure of how likely the mean is to differ from a null population, whereas the t-statistic is a measure of how big the mean difference is in units of variability.

Question 41

Suppose that the population mean for IQ is 100 with a standard deviation of 15. A researcher is testing a group 25 of talented youth. His sample has a mean of 112 with a standard deviation of 14.
What is the 95% confidence interval of the mean?

A) 106.2 - 117.8
B) 82.6 - 141.4
C) 84.56 - 139.44
D) 104.7 - 119.3

Question 42

Suppose that the population mean for IQ is 100 with a standard deviation of 15. A researcher is testing a group 25 of talented youth. His sample has a mean of 112 with a standard deviation of 14.
What does the CI tell us about the sample of young talented youth?

A) The youth have IQs comparable to that of the population
B) There is no evidence that the youth's IQ are higher than that of the general population
C) There is insufficient information to make any claims about the sample based on a 95% confidence interval.
D) The youth are indeed talented as their average IQ is significantly higher than that of the population at the alpha = .05 level

Question 43

If a one-way ANOVA has an eta squared of .24 means that…

A) the mean difference is .24 standard deviations units large.
B) the independent variable accounts for 24% of the variance in the dependent variable
C) there is a 24% chance that the independent variable caused the dependent variable
D) the independent variable and dependent variable have more than 24% overlap

Question 44

Which of the following reasons best explains the value of effect sizes?

A) Significant effects always have large effects
B) Significant effects don't always have large effects. In fact, they can have very small effects
C) Large effects tend to be significant
D) Effect sizes communicate something different than significance testing. They inform the researcher or scientist whether or not the findings can generalize to the population or not.

Question 45

Which of the following partial eta-squared is the closest to the cut off of a large effect?

A) .28
B) .40
C) .80
D) .10

Question 46

A 99% confidence interval is ____________ than a 95% confidence interval

A) less precise
B) wider
C) more likely to capture the truth
D) all of the above

Question 47

Below are the data of an course that prepares students for standardized testing. The data below show pretest (test on day 1) and a post test (test after taking the course) scores of students who enrolled. Use this data to answer the following questions.
Calculate the 95% confidence interval for the mean difference

Question 48

Below are the data of an course that prepares students for standardized testing. The data below show pretest (test on day 1) and a post test (test after taking the course) scores of students who enrolled. Use this data to answer the following questions.
Calculate cohen's d

Question 49

A research team collects data from two groups, in one group, they administered a learning task that provided feedback (i.e., telling the participant if they were right or wrong) and in the other, the participants were not provided with feedback. The research team is interested in the number of trials the participants spent learning as a function of their independent variable. Below are the number of trials each participant spent. Use it to answer the following questions.
Calculate the 95% confidence interval of the mean difference

Question 50

A research team collects data from two groups, in one group, they administered a learning task that provided feedback (i.e., telling the participant if they were right or wrong) and in the other, the participants were not provided with feedback. The research team is interested in the number of trials the participants spent learning as a function of their independent variable. Below are the number of trials each participant spent. Use it to answer the following questions.
What is the Cohen's d?

Question 51

A researcher is examining three different types of treatment on individuals who suffer from major depression. Below are the data, use it to answer the following questions
calculate the eta-squared

Question 52

A researcher is examining three different types of treatment on individuals who suffer from major depression. Below are the data, use it to answer the following questions
Characterize the size of this effect

Question 53

Use the source table below to answer the following questions
How many total eta-squared are there for this factorial ANOVA?

Question 54

Use the source table below to answer the following questions
Calculate each eta-squared

Question 55

The population average for number of children per household is 2.4 in a particular country with a standard deviation of 1.1. A research is conducting research in a neighborhood that has relatively low socioeconomic status. Her sample of 40 families averages at 3.3 children.
construct a 95% confidence interval around the sample mean.

Question 56

The population average for number of children per household is 2.4 in a particular country with a standard deviation of 1.1. A research is conducting research in a neighborhood that has relatively low socioeconomic status. Her sample of 40 families averages at 3.3 children.
what does the confidence interval tell you?