Analyzes two or more related samples or groups

Friedman Two-Way Analysis of Variance

ROXO ROCO: The research design above is:

Randomized pretest post-test with control group

Which of the following is an example of a poor experimental design?

Static-group Pretest-Posttest design

Alcohol & grades in college. Larger proportion of drinkers received low grades than others.

Drinking habits are grades are related

Results might be expressed as effect size. In these cases the differences b/w the two means is divided by:

Standard deviation of the comparison group

Which of the following is NOT possible in causal-comparative research?

random assignment to treatment or comparison groups

Rejecting the null hypothesis indicated:

Differences b/w groups is statistically significant

Which of the following results in the failure of a causal-comparative research study?

defining comparison groups carelessly

A contingency question is one that:

depends on the answer to a prior question

The difference b/w a trend study and a cohort study is that a cohort study obtains data from:

different samples from the same specific population

During the second baseline in ABA and ABAB designs, a researcher trying to show the effectiveness of an intervention

The curve reverse the trend shown in B

With every decrease of the A variable that these is a consistent increase in the B variable.

Data generate a correlation coefficient that was - # close to -1

An advantage of using interviews for survey data collection is the

opportunity to clarify questions

studbud

Research 2 Final

94 termsStudBud Administration12 plays

Quiz Review Question, Inferential Statistics PPT (#7) Approaches to Application of Statistics PPT (#8), Experimental Design PPT (#9), Single-Subject Research (#10)

Flashcards94 cardsFPractice quizPHost liveL

Questions 94

Answers

1
Which of the following represents multiple baseline observations?
C
B
A
2
Distribution of the estimates generated from many samples of the same population
Distribution of Sample Means
Distribution of Sample
Distribution of standard error
Distribution Error
3
Is another term for standard deviation, describes distribution of sample data around sample mean.
Standard Error of the Mean (SEM)
Standard Error of the Sample
Sampling error
Distribution of sample means
4
Analyzes two or more related samples or groups
Friedman Two-Way Analysis of Variance
ANCOVA
T-test for correlated proportions
Chi-Square Test
5
ROXO ROCO: The research design above is:
Randomized post-test only with control group
Randomized pretest post-test with control group
One group pretest post-test
Randomized Solomon four-group
6
Compares proportions from two different categories
t-test for independent means
T-test for proportions
Chi-Square test
ANOVA
7
T/F Inferential Statistics outcomes are hardly affected by sample size
False
True
8
Which measure of central tendency would be the best choice to summarize a clearly skewed distribution
Mean
Range
Mode
Median
9
How accurate the statistic/point estimate is from the parameter
Significance testing
Precision
Accuracy
Sample
10
T/F: Statistical Significance does indicate practical significance
True
False
11
Which of the following is an example of a poor experimental design?
randomized Solomon Group design
Static-group Pretest-Posttest design
Matching-Only design
Counterbalance
12
Alcohol & grades in college. Larger proportion of drinkers received low grades than others.
Drinking causes bad grades
Drinking habits are grades are related
Bad grades drive students to drink
Drinking habits and grades are unrelated
13
Is the magnitude of the difference or relationship big enough to matter in the real world
Practical significance
statistical significance
subjective significance
14
Results might be expressed as effect size. In these cases the differences b/w the two means is divided by:
A standard error of the mean for the experimental group
Standard deviation of the comparison group
Standard error of the mean for the comparison group
Standard deviation of the experimental group
15
One would have to pick subjects most carefully due to possible influences of subject characteristic on results?
Randomized Pretest-posttest with control group
Factorial design with randomization
Randomized Solomon Four Group
One-shot Case Study
16
This type of test is used when the area of study is better represented by the median, & for small sample sizes
Parametric
Nonparametric
17
Difference in 2 means is normal distribution, rejects null hypothesis, accept alternative hypothesis, z-scores calculate
Null hypothesis
Standard error of mean
Standing error of Differences
Standard error
18
______ research design fails to control for other variable possibly causing 'changes' observed
A-B
A-B-A
B-A-B
A-B-A-B
19
Which of the following is NOT possible in causal-comparative research?
observation
measurement
the use of comparison groups
random assignment to treatment or comparison groups
20
T/F effect size of 0.5 or larger (or -0.5 depending on direction of effect) is of interest
True
False
21
The probability of accepting the null hypothesis when it is false; failing to reject the null hypothesis that is false
Type I error
Type II error
22
T/F Unless random samples were used, interpret probabilities and significance levels as crude indices, not precise value
False
True
23
Difference between means with before and after data (paired t-test)
t-test for correlated means
ANOVA
T-test for independent means
MANOVA
24
Rejecting the null hypothesis indicated:
Differences b/w groups are due to measurement error
Differences b/w groups is statistically significant
No real statistically significant differences b/w groups
Differences b/w groups are due to chance
25
analyzes two related samples, matched on variables or tested twice
t-test for correlated means
Sign Test
Mann-Whitney U test
Chi-Square Test
26
There is a difference between the two groups, proposed explanation for an expected difference but need enough to support
Research hypothesis
Null Hypothesis
27
This type of research design allows for a little more information for interpretation of 'effect'
A-B-A
A-B-A-B
A-B-C-B
B-A-B
28
Examines the likelihood of representative hypotheses or statements about the population
Point estimate
Significance testing
Sample
Statistics
29
Which of the following results in the failure of a causal-comparative research study?
using inappropriate descriptive statistics
defining comparison groups carelessly
using groups that are too homogeneous
matching on too many variables
30
Identify the appropriate statistical procedure to investigate possible associations b/w categorical variables
scatterplot
crossbreak table
frequency polygon
correlation coefficient
31
Test when analyzing 2 or more dependent variables in same analysis. Dependent variables have degree of correlation
MANCOVA
MANOVA
t-test for independent means
ANOVA
32
Which of the following is not one of the three components of the statistical inference?
significance testing
Precision
Parameter
Point estimate
33
Which correlation Coefficient shows the strength of linear/straight-line relationship?
Pearson R
eta
34
A contingency question is one that:
Provides at least three response questions
Is contingent upon the use of specialized vocabulary
is only give to predetermined respondents
depends on the answer to a prior question
35
The difference b/w a trend study and a cohort study is that a cohort study obtains data from:
an entire population at different points in time
The same sample at different points in time
different samples from a population whose members change somewhat
different samples from the same specific population
36
During the second baseline in ABA and ABAB designs, a researcher trying to show the effectiveness of an intervention
The curve reverse the trend shown in B
The curve flatten out parallel to the x-axis
The curve continue the trend shown in B
None of these answers are correct
37
Is the index of the effect size, diff. b/w the mean of the 2 groups divided by the std. dev. of the comparison group
Magnitude
Delta
Central tendency
38
T/F: correlation always implies causation
False
True
39
Subgroup of the assessable population
Sample
Statistics
Statistical inference
Parameters
40
With every decrease of the A variable that these is a consistent increase in the B variable.
Data generate a correlation coefficient that was - # close to 0
Data generate a correlation coefficient was + close to 0
Data generate a correlation coefficient that was - # close to -1
Data generate a correlation coefficient that was + # close to 1
41
An advantage of using interviews for survey data collection is the
opportunity to clarify questions
low training requirement
low cost
anonymity of the participants
42
Type of parametric test used for categorical data
Friedman Two-Way Analysis of Variance
ANOVA
T-test for proportions
Chi-Square test
43
Standard dev. of the means from multiple from population, describes distribution of sample means around mean of means
Distribution of sample mean
Standard Error of the Mean
Sampling error
Standard error of the sample
44
Differences between means of two or more groups, usually followed by post hoc analysis if 3 or more groups
ANCOVA
MANOVA
ANOVA
t-test for independent means
45
T/F: In correlation research, the criterion variable is the variable about which the prediction is made
False
True
46
Which of the following is a nonparametric test for categorical Data?
T-test for correlated proportions
MANCOVA
Chi-Square Test
Friedman Two-Way Analysis of Variance
47
Standard deviation of the means also called standard error of mean = standard deviation of the sample means distribution
Normal distribution
Distribution of sample means
Mean of means
Sampling error
48
A negative to this type of research design is many researchers dislike leaving subjects in 'non-improved' state
A-B-A
A-B-C-B
B-A-B
A-B
49
In determining ______ - we look at the distribution of the difference between the two mean of means
SEM
null hypothesis
Probability
research hypothesis
50
confidence level (% of time population mean will be b/w 2 units of normal curve based on the SEM
SEM
stDev (of sample)
p
Standard error
51
Which type of correlation coefficient to determine the strength of curvilinear/curved relationship
Pearson R
eta
52
A causal-comparative study is used as an alternative when experimentation is:
unethical
All of these are correct
too expensive
impressive
53
Used to compare more than two independent groups again with ranking the data
T-test for independent means
ANOVA
Sign Test
Kruskal-Wallis One-Way Analysis of Variance
54
In a panel study, the researcher obtains information from:
Different samples at the same time
The same sample at different points in time
different samples at different times
An entire population at different points in time
55
Identify a disadvantage of surveys that involve direct administration to a group
researchers have no means to answer the questions
They are more expensive than other survey methods
not many types of surveys that use sample of ppl are collected as group
They produce low response rates
56
In single-subject designs, A and B refer, respectively to:
treatment and baseline
Subject A and subject B
Observer A and observer B
Baseline and treatment
57
Of the given correlation coefficients, which one represents the strongest positive relationship between variables?
0.89
0.46
0.25
0.68
58
Which of the following is a type of parametric Test for quantitative data:
Chi-Square Test
Mann-Whitney U Test
Sign Test
ANOVA
59
test that adjusts mean scores to account for initial differences between groups
ANOVA
MANCOVA
MANOVA
ANCOVA
60
Which of the following represents the treatment or intervention observations?
C
B
A
61
A comparison of delinquents vs nondelinquents might be made in order to:
test a theory on the development of delinquency
All of these answers are correct
Predict future behavior of delinquents
Determine probable factors leading to delinquency
62
Is an observed difference or relationship real based on accepted confidence levels
Practical Significance
Subjective significance
Statistical significance
63
When a cross-sectional type of data collection is undertaken that includes all ppl in the population, it is termed
trend study
cohort study
census
panel study
64
No difference or no effect between two groups, the hypothesis that is directly tested. Difference in means is 0, is =
Research hypothesis
null hypothesis
65
Which of the following is NOT a nonparametric test for quantitative data
Mann-Whitney U Test
Sign Test
Chi-Square Test
Friedman Two-Way Analysis of Varience
66
The number of data points in a condition should be:
determined by the behavior being studied
no more than five
enough to establish a stable trend
as many as is practical
67
Test for the difference between means of two samples
t-test for independent means
T-test for correlated means
MANCOVA
MANOVA
68
T/F one characteristic of experimental research is the ability of the researcher to manipulate the independent variable
True
False
69
Comparing groups for Quantitative data would include the following type of graphs except:
Frequency polygons
Histograms
Scatterplot
Box and whisker plot
70
T/F the external validity of single-subject research can be strengthened through replication studies involving others
False
True
71
____ is a technique to reduce the # of variables by searching for clusters of variables, all of which are correlated
factor analysis
Path analysis
Discriminant function analysis
Multiple regression
72
When one undertakes a statistical test with alpha equal to some number, what does this 'alpha' represent
The level of type I error that is acceptable
The degree that your sample distribution conforms to a normal distribution
The level of type II error that is acceptable
The level of both type I and type II error that is acceptable
73
To test the null hypothesis to determine the probability of obtaining a specific sample means by seeing where # falls on
Sampling error
Distribution of sample mean
Probablility
SEM
74
_____ is used to determine magnitude of difference b/w means of 2 groups regardless of statistical signficance
Central tendency
Effect size
Delta
SEM
75
Due to consideration of the _____ can be calculated w/out generating a lot of sample means and determine distribution
standard error of the mean
Standard error of sample
Distribution of sample means
Sampling error
76
The primary limitation of single-subject research is:
poor control of a maturation threat
the unreliability of measures
poor control of a history threat
limited generalizability
77
T/F: True Experimental Research designs allows the greatest control over threats to external validity
True
False
78
Same as MANOVA but with ability to account for initial differences between groups
MANCOVA
t-test for correlated means
t-test for correlation coefficient r
ANCOVA
79
Which of the following represents observations with some variation to treatment/intervention
A
C
B
80
The probability of rejecting the null hypothesis when it is in fact true
Type II error
Type I error
81
"best guess" of the population estimate
sample
Point estimate
Parameter
precision
82
Which procedures would yield the most appropriate data for studying the relationship b/w intelligence and achievement
administering achievement test to 1 sample then intelligence to another
Administering an achievement test and intelligence to one sample
Administering achievement test to 2 samples and intelligence to diff sample
Administering an achievement test to one sample of subjects, with IQ of 100
83
Which of the following is the greatest threat to internal validity in single subject research?
Morality
Subject characteristics
Instrument decay
Location
84
The importance difference between the ABA design and the ABAB design is the latter:
adds a second treatment phase
requires more time
adds a fourth data collection phase
requires shorter phases
85
Type of test that analyzes categorical data for the comparison b/w expected and actual frequencies, goodness of fit
Sign Test
T-test for proportions
Chi-Square Test
ANOVA
86
T/F Qualifications of compared groups must be random samples of specific populations.
False
True
87
The major threat to internal validity in a causal-comparative study is the possibility of a
Maturation threat
regression threat
Subject characteristics threat
History threat
88
This type of research design is typically used in situations where there insufficient time to gather baseline data
A-B
A-B-A
B-A-B
A-B-C-B
89
summarized results of all members of the pop. of interest, describing a characteristic of the pop. in numerical terms
Parameters
significance testing
statistics
sample
90
Limits within which the population mean lies
Confidence Intervals
Parameter
Statistics
Sample
91
Is similar to the t-test but has ranked (ordinal) data
Friedman Two way test
Kruskal-Wallis One-Way Analysis of Variance
Mann-Whitney U Test
Sign Test
92
An advantage of open-ended questions as compared to multiple-choice questions is that they
are preferred by respondents
provide more variety of responses
are easier to score
are less subjects to misinterpretation
93
All members of a group of interest
Sample
Point estimate
Statistics
Population
94
Tests if significantly different than 0
t-test for correlated means
ANOVA
ANCOVA
tests if significantly different than 0

Research 2 Final Flashcards | StudBud