Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
562 Cards in this Set
- Front
- Back
- 3rd side (hint)
Symbol for Null Hypothesis |
H_0 |
|
|
Null Hypothesis |
(H_0) = assumption used as the first step in statistical inference whereby the IV is said to have no effect. |
|
|
assumption used as the first step in statistical inference whereby the IV is said to have no effect.
|
Null Hypothesis |
|
|
NHST = |
Null Hypothesis Significance Testing |
|
|
Null Hypothesis Significance Testing (NHST) |
Procedure for statistical inference used to decide whether a variable has produced an effect in a study. It begins with the assumption the variable has no effect, and probability theory is used to determine the probability that the effect would occur by error variation. If the likelihood of the observed effect is small, assuming the null hypothesis is true, we infer the variable produced a reliable effect.
|
|
|
Error Variation |
chance |
|
|
Level of Significance |
The probability when testing the null hypothesis that is used to indicate whether an outcome is statistically significant. Level of significance, or alpha, is equal to the probability of a Type 1 error. |
|
|
The probability when testing the null hypothesis that is used to indicate whether an outcome is statistically significant.
|
Level of Significance |
|
|
Level of Significance is equal to ______ |
Level of significance, or alpha, is equal to the probability of a Type 1 error.
|
|
|
alpha = |
level of significance |
|
|
level of significance = |
alpha |
|
|
Type I Error |
The probability of rejecting the null hypothesis when it is true, equal to the level of significance, or alpha. |
|
|
The probability of rejecting the null hypothesis when it is true, equal to the level of significance, or alpha.
|
Type I Error |
|
|
Statistically Significant |
When the probability of an obtained difference in an experiment is smaller than would be expected if error variation alone were assumed to be responsible for the difference, the difference is statistically significant. |
|
|
When the probability of an obtained difference in an experiment is smaller than would be expected if error variation alone were assumed to be responsible for the difference, the difference is _____________
|
Statistically Significant |
|
|
Type II Error |
The probability of failing to reject the null hypothesis when it is false. |
|
|
The probability of failing to reject the null hypothesis when it is false.
|
Type II Error |
|
|
Validity |
the "truthfulness" of a measure; one that measures what it claims to measure. |
|
|
A measure that measures what it claims to measure is ____________ |
Valid |
|
|
Threats to Internal Validity |
Possible causes that must be controlled so a clear cause-effect inference can be made. |
|
|
These must be controlled for if we want to be able to make a clear cause-effect inference. |
Threats to Internal Validity |
|
|
Scatterplot |
graph showing the relationship between two variables by indicating the intersection of two measures obtained from the same person, thing, or event. |
|
|
graph showing the relationship between two variables by indicating the intersection of two measures obtained from the same person, thing, or event.
|
Scatterplot |
|
|
Sample |
something less than all the cases of interest; in survey research, a subset of the research |
|
|
something less than all the cases of interest; in survey research, a subset of the research
|
Sample |
|
|
ABAB Design = |
Reversal Design |
|
|
Reversal Design = |
ABAB Design |
|
|
Reliability |
when a measurement is consistent |
|
|
When a measurement is consistent it is _________ |
Reliable |
|
|
Probability Sampling |
Sampling procedure in which the probability that each element of the population will be included in the sample can be specified. |
|
|
Sampling procedure in which the probability that each element of the population will be included in the sample can be specified.
|
Probability Sampling |
|
|
Practice Effects |
Changes that participants undergo with repeated testing. They are the summation of both positive ( ex - familiarity with a test) and negative (ex - boredom) factors associated with repeated measures. |
|
|
Changes that participants undergo with repeated testing. They are the summation of both positive ( ex - familiarity with a test) and negative (ex - boredom) factors associated with repeated measures.
|
Practice Effects |
|
|
Power |
Probability in a statistical test that a false null hypothesis will be rejected; power is related to:
|
S.E.A. |
|
Probability in a statistical test that a false null hypothesis will be rejected; It is related to the level of significance selected, the size of the treatment effect, and the sample size.
|
Power |
|
|
Power is related to ______________ |
|
S.E.A. |
|
Probability in a statistical test that a false null hypothesis will be rejected
|
Power |
|
|
ABAB Design |
(AKA - REVERSAL DESIGN) |
|
|
Alternative Hypothesis = |
H_1 |
|
|
Alternative Hypothesis |
The hypothesis that states a treatment *does* have an effect. |
|
|
The Hypothesis that states a treatment *does* have an effect. |
Alternative Hypothesis (H_1) |
|
|
The probability of getting a particular effect if the H_0 were true.
|
P-Value |
|
|
P-Value |
The probability of getting a particular effect if the H_0 were true. |
|
|
Effect Size |
Index of the strength of the relationship between the independent variable and dependent variable that is independent of the sample size. |
|
|
Index of the strength of the relationship between the independent variable and dependent variable that is independent of the sample size.
|
Effect Size |
|
|
_______ Is based on probability. |
NHST (This means it can support a hypothesis, but never prove or disprove one; there is always a possibility for error)
|
|
|
The probability of obtaining the effect you got if the null hypothesis were true. |
P-Value |
|
|
When should you reject, or fail to reject a null hypothesis based on a given P-Value |
if the p-value is less than or equal to the significance level of the test we reject the null and conclude the alternate hypothesis is true. If the p-value is greater than the significance level then we fail to reject the null hypothesis and conclude it is plausible.
|
|
|
Cohen's d |
THE EFFECT SIZE OF THE DIFFERENCE BETWEEN GROUPS. Based on Cohen's guidelines d values of . 20 = small effect, .30 = medium effect, .80 = large effect of an independent variable. |
|
|
d-value of .20 = |
small effect size if an independent variable (Cohen's d) |
|
|
d-value of .80 = |
large effect size of an independent variable (Cohen's d) |
|
|
small effect size of an independent variable is = to a Cohen's d of ______ |
.20 |
|
|
a medium effect size of an independent variable is equal to a Cohen's d of _____ |
.30 |
|
|
Dependent Variable |
A measure of behavior used to assess the effect of the independent variable. |
|
|
A measure of behavior used to assess the effect of the independent variable |
Dependent Variable |
|
|
Independent Variable |
Factor for which the researcher manipulates at least two levels in order to determine its effect on behavior |
|
|
Factor for which the researcher manipulates at least two variables in order to determine its effect on behavior. |
Independent Variable |
|
|
MEASURES THE EFFECT SIZE (STRENGTH) OF A CORRELATION.
|
Pearson's r |
|
|
Pearson's r |
MEASURES THE EFFECT SIZE (STRENGTH) OF A CORRELATION
|
|
|
More ____________ makes it more likely you will detect an effect. |
Power |
|
|
More power makes it more likely that you will |
Detect an effect |
|
|
How is an experiment used to infer causality? |
|
|
|
Control requires _____________ |
balanced samples |
|
|
__________ requires balanced samples |
control |
|
|
What do you need in order to have a causal inference? |
|
|
|
Time-Order Relationship |
The IV MUST preceed the DV. |
|
|
The IV MUST preceed the DV. |
Time-Order Relationship |
|
|
Time-Order Relationship is necessary in order to ________________ |
infer causality |
|
|
Co-variation |
Performance on the DV changes for different levels of the IV (ex - exam score differs for those who had caffeine compared to those who do not.) |
|
|
When performance on the DV changes for different levels of the IV. |
Co-variation |
|
|
How do we eliminated plausible alternative explanations for a causal inference? |
Elimination of plausible alternative explanations is only achieved through balanced groups. |
|
|
Why are experiments the best way to determine causality? |
Because if control is sufficient and all causal requirements are met, then any differences between the levels of the IV must be caused by the independent variable. |
|
|
degree to which differences in performance can be attributed unambiguously to an effect of an IV, as opposed to an effect of some other uncontrolled variable
|
Internal Validity |
|
|
A study that is _______________ is free from confounds. |
Internally Valid |
|
|
Attrition is a threat to _______ |
Internal Validity |
|
|
Subject Attrition AKA |
Attrition |
|
|
Attrition |
a threat to internal validity that occurs when participants are lost from an experiment, for example, when participants drop out of the research. The loss of participants changes the nature of a group from that established prior to the introduction of the treatment. (Ex - by destroying the equivalence of groups that had been established through random assignment.) |
|
|
a threat to internal validity that occurs when participants are lost from an experiment, for example, when participants drop out of the research. The loss of participants changes the nature of a group from that established prior to the introduction of the treatment. (Ex - by destroying the equivalence of groups that had been established through random assignment.)
|
Attrition |
|
|
Selective Subject Loss |
Occurs when subjects are lost differentially across the conditions of the experiment as the result of some characteristic of each subject that is related to the outcome of the study. |
|
|
True or False: Selective Subject Loss is not a big deal |
False. |
|
|
The extent to which the results of a study can be generalized to different populations, settings, and conditions.
|
External Validity |
|
|
Both the participant and the observer are kept unaware of what treatment is being administered.
|
Double-Blind Procedure |
|
|
Independent Groups Design |
Each seperate group of subjects represents a different condition as defined by the IV. |
|
|
Design where each seperate group of subjects represents a different condition as defined by the IV.
|
Independent Groups Design |
|
|
Matched-Groups Design |
Type of Independent Groups Design in which the researcher forms comparable groups by matching subjects on a pretest task and then randomly assigning the members of those matched sets of subjects to the conditions of the experiment. |
|
|
Type of Independent Groups Design in which the researcher forms comparable groups by matching subjects on a pretest task and then randomly assigning the members of those matched sets of subjects to the conditions of the experiment.
|
Matched-Groups Design |
|
|
Natural Groups Deign |
Type of independent groups design in which the conditions represent the selected levels of a naturally occurring independent variable, for example, the individual differences variable age. |
|
|
Type of independent groups design in which the conditions represent the selected levels of a naturally occurring independent variable, for example, the individual differences variable age.
|
Natural Groups Design |
|
|
Three different types of independent groups designs |
|
|
|
Random, Natural, and Matched are what kind of groups designs? |
Independent Group Designs |
|
|
Repeated Measures Design |
Research designs in which each subject participates in all conditions of the experiment (ex - all measurement is repeated on the same subject.) |
|
|
Research designs in which each subject participates in all conditions of the experiment
|
Repeated Measures Design |
|
|
What type design experiences a unique confound called Practice Effect? |
Repeated Measures Designs |
|
|
Counterbalancing |
Control technique for distributing (balancing) practice effects across the conditions of a repreated measures design. How it is accomplished depends on whether or not the study is a complete or incomplete repeated measures design. |
|
|
Control technique for distributing (balancing) practice effects across the conditions of a repreated measures design. How it is accomplished depends on whether or not the study is a complete or incomplete repeated measures design.
|
Counterbalancing |
|
|
How do you figure out how many orders would be required to counterbalance with an All Possible Orders form of Counterbalancing? |
N! |
|
|
Differential Transfer |
Potential problem in repeated measures designs when performance in one condition differs depending on which of two other conditions preceeds it. |
|
|
Potential problem in repeated measures designs when performance in one condition differs depending on which of two other conditions preceeds it.
|
Differential Transfer |
|
|
Source of evidence based on records or documents relating the activities of individuals, institutions, governments, and other groups
|
Archival Data |
|
|
MEASURES OF BEHAVIOR THAT ELIMINATE THE PROBLEM OF REACTIVITY BECAUSE OBSERVATIONS ARE MADE IN SUCH A WAY THAT THE PRESENCE OF THE OBSERVER IS NOT DETECTED BY THOSE BEING OBSERVED.
|
UNOBTRUSIVE (NON-REACTIVE) MEASURES
|
|
|
_____________ USE REAL WORLD EVIDENCE TO TEST HYOTHESES WITHOUT PARTICIPANT INVOLVEMENT. |
NON-REACTIVE MEASURES |
|
|
SOURCE OF THE EVIDENCE THAT IS BASED ON THE REMNANTS, FRAGMENTS, AND PRODUCTS OF PAST BEHAVIOR
|
PHYSICAL TRACES |
|
|
USE TRACES |
PHYSICAL EVIDENCE THAT RESULTS FROM USE |
|
|
USE TRACES THAT COME FROM NATURALLY OCCURING EVENTS |
NATURAL USE TRACES |
|
|
CONTROLLED USE TRACES |
INVOLVE MANIPULATION BY THE RESEARCHER |
|
|
PRODUCTS |
ARTIFACTS THAT GIVE YOU INFO ABOUT A PERSON OR CULTURE |
|
|
ARTIFACTS THAT GIVE YOU INFO ABOUT A PERSON OR CULTURE. |
PRODUCTS |
|
|
TYPES OF ARCHIVAL DATA RECORDS |
RUNNING RECORDS MEDIA NATURAL TREATMENTS |
|
|
RUNNING RECORDS, MEDIA, AND NATURAL TREATMENTS ARE TYPES OF WHAT SORT OF DATA RECORDS? |
ARCHIVAL DATA |
|
|
RUNNING RECORDS |
RECORDS THAT ARE CONTINUALLY KEPT AND UPDATED |
|
|
RECORDS THAT ARE CONTINUALLY KEPT AND UPDATED |
RUNNING RECORDS |
|
|
TYPE OF ARCHIVAL DATA RECORDS THAT TRACKS EFFECTS OF SOCIETAL EVENTS |
NATURAL TREATMENTS |
|
|
NATURAL TREATMENTS ARE |
TYPES OF ARCHIVAL DATA RECORDS THAT TRACK EFFECTS OF SOCIETAL EVENTS. |
|
|
TRUE OR FALSE: ARCHIVAL DATA CAN NOT REFER TO DATA COLLECTED IN PREVIOUS STUDIES. |
FALSE, DATA COLLECTED IN PREVIOUS STUDIES CAN BE CONSIDERED ARCHIVAL DATA. |
|
|
ARCHIVAL DATA RECORDS THAT CONSIST OF NEWS REPORTS, BOOKS, MOVIES, ADVERTISEMENTS
|
MEDIA RECORDS |
|
|
TWO TYPES OF ARCHIVAL DATA STUDIES |
META-ANALYSIS SECONDARY-DATA ANALYSIS |
|
|
META-ANALYSIS IS WHAT KIND OF STUDY? |
ARCHIVAL DATA STUDY |
|
|
META-ANALYSIS |
A STUDY OF PREVIOUS STUDIES THAT COMBINES AND ANALYZES DATA FROM MANY STUDIES ON A TOPIC TO MAKE MORE DEFINITIVE CONCLUSIONS |
|
|
SECONDARY DATA ANALYSIS |
A TYPE OF ARCHIVAL DATA STUDY THAT ANALYZES PREVIOUSLY COLLECTED DATA FOR A NEW PURPOSE. |
|
|
A TYPE OF ARCHIVAL DATA STUDY THAT ANALYZES PREVIOUSLY COLLECTED DATA FOR A NEW PURPOSE.
|
SECONDARY DATA ANALYSIS |
|
|
SELECTIVE DEPOSIT |
WHEN ARCHIVAL/TRACE SOURCES ARE INCOMPLETE OR INACCURATE BECAUSE SOME ARE MORE LIKELY TO BE CREATED THAN OTHERS. |
|
|
SELECTIVE SURVIVIAL |
SOME ARCHIVES OR TRACES MAY SURVIVE OVER TIME WHERE OTHERS HAVE NOT, THUS THEY MAY NOT BE AN ACCURATE MEASURE. |
|
|
SOME ARCHIVAL/TRACES MAY SURVIVE OVER TIME WHERE OTHERS HAVE NOT; AND THUS MAY NOT BE AN ACCURATE MEASURE. |
SELECTIVE SURVIVAL |
|
|
SELECTION BIAS |
THREAT TO THE REPRESENTATIVENESS OF A SAMPLE THAT OCCURS WHEN THE PROCEDURES USED TO SELECT A SAMPLE RESULT IN THE OVER OR UNDER REPRESENTATION OF A SIGNIFICANT SEGMENT OF THE POPULATION. |
|
|
THREAT TO THE REPRESENTATIVENESS OF A SAMPLE THAT OCCURS WHEN THE PROCEDURES USED TO SELECT A SAMPLE RESULT IN THE OVER OR UNDER REPRESENTATION OF A SIGNIFICANT SEGMENT OF THE POPULATION.
|
SELECTION BIAS |
|
|
TYPES OF SELECTION BIASES |
SELECTIVE DEPOSIT AND SELECTIVE SURVIVAL |
|
|
IN ARCHIVAL ANALYSIS _______________ LIMITS THE GENERALITY OF RESEARCH FINDINGS |
SELECTIVE DEPOSIT |
|
|
IN ARCHIVAL ANALYSIS _____________ LIMITS THE EXTERNAL VALIDITY OF RESEARCH FINDINGS |
SELECTIVE SURVIVAL |
|
|
ADVANTAGES OF ARCHIVAL DATA ANALYSIS |
|
|
|
DISADVANTAGES OF ARCHIVAL DATA ANALYSIS |
|
|
|
ALL RESPONDENDS COMPLETE THE SAME ITEMS, VERBALLY (INTERVIEW) OR IN WRITING (QUESTIONNAIRE)
|
SURVEY |
|
|
A WRITTEN SURVEY IS A ___________ |
QUESTIONNAIRE |
|
|
A VERBAL SURVEY IS A ____________ |
INTERVIEW |
|
|
ANSCOMB'S QUARTET |
comprises four datasets that have nearly identical simple statistical properties, yet appear very different when graphed. Each dataset consists of eleven (x,y) points. They were constructed in 1973 by the statistician Francis Anscombe to demonstrate both the importance of graphing data before analyzing it and the effect of outliers on statistical properties
|
|
|
comprises four datasets that have nearly identical simple statistical properties, yet appear very different when graphed. Each dataset consists of eleven (x,y) points. They were constructed to demonstrate both the importance of graphing data before analyzing it and the effect of outliers on statistical properties
|
ANSCOMB'S QUARTET |
|
|
CORRELATION |
EXISTS WHEN TWO DIFFERENT MEASURES OF THE SAME PEOPLE, EVENTS, OR THINGS VARY TOGETHER; THE PRESENCE OF A CORRELATION MAKES IT POSSIBLE TO PREDICT VALUES ON ONE VARIABLE BY KNOWING THE VALUES ON A SECOND VARIABLE. |
|
|
EXISTS WHEN TWO DIFFERENT MEASURES OF THE SAME PEOPLE, EVENTS, OR THINGS VARY TOGETHER
|
CORRELATION |
|
|
THE PRESENCE OF A CORRELATION MAKES IT POSSIBLE TO
|
PREDICT VALUES ON ONE VARIABLE BY KNOWING THE VALUES ON A SECOND VARIABLE.
|
|
|
REPRESENTATIVENESS |
A SAMPLE IS REPRESENTATIVE TO THE EXTENT THAT IT HAS THE SAME DISTRIBUTION OF CHARACTERISTICS AS THE POPULATION FROM WHICH IT WAS SELECTED; ABILITY TO GENERALIZE FROM A SAMPLE IS CRITICALLY LINKED TO ITS REPRESENTATIVENESS. |
|
|
ABILITY TO ___________ FROM A SAMPLE IS CRITICALLY LINKED TO ITS REPRESENTATIVENESS.
|
GENERALIZE |
|
|
POPULATION |
SET OF ALL CASES OF INTEREST |
|
|
SET OF ALL CASES OF INTEREST |
POPULATION |
|
|
SAMPLING FRAME |
SPECIFIC LIST OF A SUBSET OF THE POPULATION; IN A SENSE THE OPERATIONAL DEFINITION OF THE POPULATION. |
|
|
THE SUBSET OF THE POPULATION DRAWN FROM THE SAMPLING FRAME THAT IS INCLUDED IN THE STUDY |
SAMPLE |
|
|
ELEMENT IN SAMPLING IS A |
INDIVIDUAL MEMBER OF A POPULATION |
|
|
IN SAMPLING AN INDIVIDUAL MEMBER OF A POPULATION IS KNOWN AS AN |
ELEMENT |
|
|
SELECTION BIAS IS A THREAT TO THE _________ OF A SAMPLE |
REPRESENTATIVENESS |
|
|
PROBABILITY SAMPLING |
SAMPLING PROCEDURE IN WHICH THE PROBABILITY THAT EACH ELEMENT OF THE POPULATION WILL BE INCLUDED IN THE SAMPLE CAN BE SPECIFIED. |
|
|
SAMPLING PROCEDURE IN WHICH THE PROBABILITY THAT EACH ELEMENT OF THE POPULATION WILL BE INCLUDED IN THE SAMPLE CAN BE SPECIFIED.
|
PROBABILITY SAMPLING |
|
|
SIMPLE RANDOM SAMPLING |
EACH ELEMENT HAS THE SAME PROBABILITY OF INCLUSION. |
|
|
TYPE OF PROBABILITY SAMPLING WHEREIN EACH ELEMENT HAS THE SAME PROBABILITY OF INCLUSION. |
SIMPLE RANDOM SAMPLING |
|
|
STRATIFIED RANDOM SAMPLING |
POPULATION IS DIVIDED INTO STRATIFIED SUBPOPULATIONS AND RANDOM SAMPLES ARE DRAWN FROM EACH OF THE STRATA. |
|
|
POPULATION IS DIVIDED INTO STRATIFIED SUBPOPULATIONS AND RANDOM SAMPLES ARE DRAWN FROM EACH
|
STRATIFIED RANDOM SAMPLING |
|
|
WHAT MAKES A SURVEY OR POLL SCIENTIFIC? |
GOOD REPRESENTATIVENESS |
|
|
NON-PROBABILITY SAMPLING |
A SAMPLE IS NOT RANDOM, USUALLY DONE OUT OF CONVIENENCE, BRONE TO BIASES. |
|
|
TYPE OF SAMPLING USED IN SCIENTIFIC POLLS |
PROBABILITY SAMPLING |
|
|
RESPONSE BIAS |
WHEN SOME PEOPLE ARE MORE LIKELY TO RESPOND TO SURVEYS THAN OTHERS. |
|
|
BIAS ENCOUNTERED WHEN SOME PEOPLE ARE MORE LIKELY TO RESPOND TO SURVEYS THAN OTHERS. |
RESPONSE BIAS. |
|
|
CROSS-SECTIONAL DESIGN |
SURVEY RESEARCH DESIGN WHERE ONE OR MORE SAMPLES OF THE POPULATION ARE SELECTED AND INFO IS COLLECTED FROM THE SAMPLES AT ONE TIME. |
|
|
SURVEY RESEARCH DESIGN WHERE ONE OR MORE SAMPLES OF THE POPULATION ARE SELECTED AND INFO IS COLLECTED FROM THE SAMPLES AT ONE TIME.
|
CROSS-SECTIONAL DESIGN |
|
|
SUCCESSIVE-INDEPENDENT SAMPLES DESIGN |
SURVEY RESEARCH DESIGN WHERE A SERIES OF CROSS-SECTIONAL SURVEYS IS DONE AND THE SAME QUESTIONS ARE ASKED OF EACH SUCCEEDING SAMPLE OF RESPONDENTS. |
|
|
SURVEY RESEARCH DESIGN WHERE A SERIES OF CROSS-SECTIONAL SURVEYS IS DONE AND THE SAME QUESTIONS ARE ASKED OF EACH SUCCEEDING SAMPLE OF RESPONDENTS.
|
SUCCESSIVE-INDEPENDENT SAMPLES DESIGN |
|
|
LONGITUDINAL DESIGN |
RESEARCH DESIGN WHERE THE SAME SAMPLE OF RESPONDENTS IS INTERVIEWED OR TESTED MORE THAN ONCE. |
|
|
RESEARCH DESIGN WHERE THE SAME SAMPLE OF RESPONDENTS IS INTERVIEWED OR TESTED MORE THAN ONCE.
|
LONGITUDINAL DESIGN |
|
|
PROBLEM WITH CROSS-SECTIONAL DESIGNS |
IT'S A SNAPSHOT IN TIME, YOU CAN'T ASSESS CHANGE OVER TIME. |
|
|
PROBLEMS WITH SUCCESSIVE INDEPENDENT SAMPLES (COHORTS) |
QUESTIONS AND SAMPLING MUST REMAIN CONSISTENT. |
|
|
PROBLEMS WITH LONGITUDINAL DESIGNS |
SAMPLE ATTRITION |
|
|
THREE TYPES OF RELIABILITY |
|
|
|
INTERNAL CONSISTENCY |
DO ALL THE QUESTIONS/ITEMS MEASURE THE SAME THING? |
|
|
IF ALL THE QUESTIONS AND ITEMS ON A TEST MEASURE THE SAME THING WE SAY IT HAS _______ |
INTERNAL CONSISTENCY |
|
|
TEST-RETEST RELIABILITY |
A FORM OF RELIABILITY WHERE EACH OF THE ITEMS MEASURE THE SAME THING EACH TIME. |
|
|
A FORM OF RELIABILITY WHERE EACH OF THE ITEMS MEASURE THE SAME THING EACH TIME.
|
TEST-RETEST RELIABILITY |
|
|
FOUR TYPES OF VALIDITY |
|
F.C.D.C |
|
FACE VALIDITY |
IS IT OBVIOUS WHAT THE ITEMS ARE INTENDED TO MEASURE |
|
|
IF IT IS OBVIOUS WHAT THE ITEMS ARE INTENDED TO MEASURE WE SAY IT HAS _______ |
FACE VALIDITY |
|
|
CONVERGENT VALIDITY |
IS THE MEASURE CORRELATED WITH VALID MEASURES OF THE SAME CONSTRUCT? |
|
|
IF THE MEASURE CORRELATES WITH VALID MEASURES OF THE SAME CONSTRUCT WE SAY IT HAS |
CONVERGENT VALIDITY |
|
|
DISCRIMINANT VALIDITY |
DOES IT DISTINGUISH BETWEEN GROUPS? |
|
|
IF THE MEASURE IS ASSOCIATED WITH REAL WORLD EXAMPLES OF THE CONSTRUCT WE SAY IT HAS |
CRITERION-PREDICTION VALIDITY |
|
|
Cronbach's alpha |
is used as an estimate of the reliability of a test. It can be viewed as the expected correlation of two tests that measure the same thing. |
|
|
is used as a (lowerbound) estimate of the reliability of a psychometric test. It has been proposed that can be viewed as the expected correlation of two tests that measure the same construct.
|
CRONBACH'S ALPHA |
|
|
THE PROBABILITY THAT A FALSE H_0 WILL BE REJECTED. |
POWER |
|
|
Variability AKA |
MEASURES OF DISPERSION |
|
|
MEASURES OF DISPERSION AKA |
VARIABILITY |
|
|
MEASURES OF DISPERSION |
MEASURES LIKE RANGE AND STD. DV8 THAT DESCRIBE THE DEGREE OF DISPERSION OF NUMBERS IN A DISTRIBUTION. |
|
|
MEASURES LIKE RANGE AND STD. DV8 THAT DESCRIBE THE DEGREE OF DISPERSION OF NUMBERS IN A DISTRIBUTION.
|
MEASURES OF DISPERSION |
|
|
range |
the difference between the highest and lowest score in a distribution. highest-lowest=range |
|
|
the difference between the highest and lowest score in a distribution.
|
range |
|
|
standard deviation |
the most commonly used measure of dispersion that indicates approximately how far on the average scores differ from the mean. |
|
|
the most commonly used measure of dispersion that indicates approximately how far on the average scores differ from the mean.
|
standard deviation |
|
|
standard deviation is a measure of __________ |
dispersion. It shows how far (on average) scores differ from the mean. |
|
|
Inferential Statistics |
makes inferences about populations using data drawn from the population. Instead of using the entire population to gather the data, the statistician will collect a sample or samples from the millions of residents and make inferences about the entire population using the sample.
|
|
|
makes inferences about populations using data drawn from the population. Instead of using the entire population to gather the data, the statistician will collect a sample or samples from the millions of residents and make inferences about the entire population using the sample.
|
inferential statistics |
|
|
If an effect is unlikely to have occurred by chance it is |
statistically significant |
|
|
Statistically Significant |
When an event is unlikely to have occured by chance. |
|
|
What does Null Hypothesis Significance Testing show us? |
It is the likelihood that the difference we see in our groups occurred by chance alone and not because of the IV |
|
|
Uses statistics to determine if variable relationships from a sample reflect true relationships in the general population.
|
Null Hypothesis Significance Testing (NHST) |
|
|
the likelihood that the difference we see in our groups occurred by chance alone and not because of the IV
|
NHST -Null Hypothesis Significance Testing |
|
|
|
Null Hypothesis H_0 |
|
|
|
Alternative Hypothesis H_1 |
|
|
alpha level |
the p-value threshold that needs to be crossed for 'statistical significance' (typically p < .05 or less than 5%) |
|
|
the p-value threshold that needs to be crossed for 'statistical significance' (typically p < .05 or less than 5%)
|
alpha level |
|
|
What number in an alpha level indicates statistical significance? |
p < .05 (less than 5%) |
|
|
TRUE OR FALSE: YOU DON'T NEED TO REPORT THE EXACT P-VALUE FOR A STUDY |
FALSE, YOU SHOULD REPORT THE EXACT P-VALUE |
|
|
AN ALPHA LEVEL OF P < .10 INDICATES |
A TREND TOWARD SIGNIFICANCE |
|
|
WHEN SHOULD YOU CHOOSE YOUR ALPHA LEVEL FOR A STUDY? |
BEFORE YOU BEGIN IN ORDER TO CONTROL FOR EXPERIMENTER BIAS |
|
|
WHEN SHOULD YOU PICK A LOWER ALPHA LEVEL FOR YOUR STUDY? |
IF YOU ARE CONDUCTING MANY ANALYSES. |
|
|
IN ORDER TO AVOID INFLATING RESULTS IN A STUDY THAT CONTAINS MANY ANALYSES, WHAT ALPHA LEVEL SHOULD YOU USE? |
A LOWER VALUE (.01 OR .001) |
|
|
TRUE OR FALSE: WITH INFERENTIAL STATISTICS YOU CAN RELY EXCLUSIVELY ON THE MEAN. |
FALSE, YOU ALSO NEED TO CONSIDER VARIABILITY. |
|
|
WHAT DOES IT MEAN TO REJECT THE NULL HYPOTHESIS? |
IT MEANS THERE WAS AN ALPHA LEVEL OF P < .05 (OR WHATEVER WAS SELECTED BEFORE THE STUDY) AND THAT IT IS UNLIKELY THE RESULTS OBSERVED HAPPENED BY CHANCE. THIS MEANS THE IV DID HAVE AN IMPACT ON THE DV AND THAT THE DATA SUPPORTS THE H_1 (ALTERNATIVE HYPOTHESIS) |
|
|
TRUE OR FALSE: WHEN YOU REJECTING OR FAILING TO REJECT THE NULL HYPOTHESIS PROVES EITHER THE H_0 OR H_1 |
FALSE: THERE IS STILL SOME CHANCE THE RESULTS COULD BE RANDOM OR DUE TO SOME CONFOUNDING FACTOR. YOU CAN ONLY *SUPPORT* A HYPOTHESIS OR NULL HYPOTHESIS, YOU CAN'T PROVE OR DISPROVE IT. |
|
|
IF GIVEN AN ALPHA LEVEL WHERE THE P-VALUE IS GREATER THAN .05 SHOULD YOU REJECT, OR FAIL TO REJECT THE H_0? |
FAIL TO REJECT THE H_0 |
|
|
IF GIVEN AN ALPHA LEVEL WHERE THE P-VALUE IS LESS THAN .05 SHOULD YOU REJECT, OR FAIL TO REJECT THE H_0?
|
REJECT THE H_0 |
|
|
AN ALPHA LEVEL OF P < .05 SUPPORTS THE |
H_1 OR ALTERNATIVE HYPOTHESIS |
|
|
AN ALPHA LEVEL OF P > .05 SUPPORTS THE
|
NULL HYPOTHESIS (H_0) |
|
|
TESTS THE PROBABILITY THAT THE NULL HYPOTHESIS IS TRUE GIVEN THE PATTERNS FROM OUR DATA |
NHST |
|
|
THE PROBABILITY OF GETTING THOSE RESULTS IF THE NULL HYPOTHESIS WERE TRUE |
P-VALUE |
|
|
THRESHOLD THAT THE P-VALUE NEEDS TO BE UNDER IN ORDER TO REJECT THE NULL HYPOTHESIS |
ALPHA LEVEL |
|
|
TYPE OF ERROR THAT IS A FALSE POSITIVE |
TYPE 1 |
|
|
TYPE OF ERROR THAT IS A FALSE NEGATIVE |
TYPE 2 |
|
|
REJECTING THE NULL HYPOTHESIS WHEN IT WAS ACTUALLY TRUE |
TYPE 1 ERROR |
|
|
FAILING TO REJECT THE NULL HYPOTHESIS WHEN IT WAS ACTUALLY FALSE |
TYPE 2 ERROR |
|
|
THIS TYPE OF ERROR MOST COMMONLY OCCURS WHEN RUNNING MANY ANALYSES |
TYPE 1 ERROR |
|
|
TYPE 1 ERROR MOST COMMONLY OCCURS WHEN |
RUNNING MANY ANALYSES |
|
|
TO REDUCE THE CHANCES OF A TYPE 1 ERROR YOU SHOULD |
HAVE A MORE STRICT P-VALUE |
|
|
HAVING A MORE STRICT P-VALUE REDUCES THE CHANCE OF WHAT TYPE OF ERROR |
TYPE 1 ERROR (FALSE POSITIVE) |
|
|
TYPE 2 ERROR IS USUALLY CAUSE BY |
A SMALL SAMPLE SIZE |
|
|
TYPE OF ERROR THAT IS GENERALLY CAUSED BY TOO SMALL A SAMPLE SIZE |
TYPE 2 ERROR |
|
|
A P-VALUE IS DEPENDENT ON ______
|
SAMPLE SIZE |
|
|
PEARSON'S R AND COHEN'S D ARE TYPES OF |
EFFECT SIZES |
|
|
THE EFFECT SIZE OF A CORRELATION |
PEARSON'S R |
|
|
PEARSON'S R MEASURES |
THE EFFECT SIZE OF A CORRELATION |
|
|
PEARSON'S R IS EQUAL TO |
THE STRENGTH OF THE ASSOCIATION BETWEEN TWO VARIABLES |
|
|
MEASURES THE STRENGTH OF THE CORRELATION BETWEEN VARIABLES |
PEARSON'S R |
|
|
A PEARSON'S R = TO 0 - .10 MEANS |
THERE IS NO CORRELATION BETWEEN THE VARIABLES. |
|
|
IN ORDER FOR THERE TO BE NO CORRELATION BETWEEN THE VARIABLES YOU NEED A PEARSON'S R OF |
R = TO 0 - .10 |
|
|
A PEARSON'S R > .10 MEANS |
THE CORRELATION BETWEEN VARIABLES IS SMALL |
|
|
PEARSON'S R VALUE THAT INDICATES A SMALL CORRELATION BETWEEN THE VARIABLES |
R > .10 |
|
|
PERSON'S R > .37 MEANS |
THERE IS A STRONG CORRELATION BETWEEN THE VARIABLES |
|
|
PEARSON'S R LEVELS FOR PSYCHOLOGY |
NONE = R = 0 - .10 SMALL = R > .10 MEDIUM = R > .24 LARGE = R > .37 |
|
|
WHAT IS USED TO DETERMINE COHEN'S D |
2. THE AMOUNT OF VARIABILITY BETWEEN THE GROUPS |
|
|
WHAT ARE EFFECT SIZES USED FOR? |
|
|
|
THE ABILITY TO DETECT STATISTICALLY SIGNIFICANT EVENTS |
POWER |
|
|
1 - TYPE 2 ERROR = |
POWER (WE TYPICALLY WANT OVER .80.) |
|
|
1 - ______= POWER |
TYPE 2 ERROR |
|
|
LEVEL OF POWER WE GENERALLY WANT IN A STUDY |
.80 OR GREATER |
|
|
A SMALLER ____ LEVEL MEANS LESS POWER |
ALPHA LEVEL |
|
|
WHY IS EFFECT SIZE A FACTOR IN POWER? |
BECAUSE IT IS EASIER TO DETECT A LARGE EFFECT THAN A SMALL ONE. |
|
|
WHY DO LARGER SAMPLE SIZES GIVE YOU MORE POWER? |
BECAUSE IT IS A BETTER ESTIMATE OF THE POPULATION AND YOU ARE MORE LIKELY TO DETECT A STATISTICALLY SIGNIFICANT EVENT IF ONE EXISTS. |
|
|
Why should you be careful not to have too much power in a study? |
Because with a large enough sample size any difference at the population level will be statistically significant. This increases the chances of a type 1 error. |
|
|
A higher than average p-value can increase the likelihood of ____ error |
Type 1 error |
|
|
Is type 1 error or type 2 error the most serious? |
Type 1 is more serious, but type 2 is more common. |
|
|
Which is more common, type 1 or type 2 errors? |
Type 2 is more common, but type 1 is more serious. |
|
|
A power analysis helps us determine |
the necessary sample size |
|
|
What do you need to know in order to conduct a power analysis? |
Effect Size of Interest and the planned p-value you want to obtain |
|
|
_________ tells you the sample size you need for the effect size to reach that level of statistical significance. |
a power analysis |
|
|
Descriptive designs do what? |
measure, but do not manipulate variables |
|
|
drawbacks of descriptive designs measures |
Descriptive Designs measure, but don't manipulate the variables (This means they lack control). Because of this we can establish correlations, but not infer causality |
|
|
1 Group Pre-Test-Post-Test Design |
a single group is tested before and after treatment. |
|
|
Design where a single group is tested before and after treatment |
1 Group Pre-Test-Post-Test Design |
|
|
is a 1 Group Pre-Test-Post-Test Design an experiment? |
No because it has manipulation, but no control and it fails to eliminate all other explanations for the results |
|
|
The extent to which you can make a causal inference based on the experiment |
Internal Validity |
|
|
Internal Validity |
The extent to which you can make a causal inference based on the experiment |
|
|
Intact Groups |
using pre-existing groups |
|
|
Independent Groups Design |
Each group in the experiment represents a different level of the IV |
|
|
This type of design examines between group differences. |
Independent Groups Design |
|
|
Independent Groups Design examines what kind of group differences? |
Between Group Differences |
|
|
What are Between Group Differences? |
Between group differences are the effect of the IV on the DV. |
|
|
Within Group Differences |
Within group differences are individual differences or error. |
|
|
Best Variables For Matched Groups |
1) The DV 2) Related to the DV 3) A Potential Confound |
|
|
Why can't natural groups use random assignment? |
Because it involves naturally occuring IVs that you cannot ethically or practically manipulate. |
|
|
Why can't matched groups use random assignment? |
Because the groups are too small to balance out the individual differences. |
|
|
How do we infer causality with natural groups? |
introduce another IV that *can* be manipulated by the experimenter. |
|
|
Experimental Design for which practice effects are a concern |
Repeated Measures Design |
|
|
What Experimental Design should you use if you want to use fewer participants than Individual Groups Designs? |
Repeated Measures Groups |
|
|
How do you control for practice effects? |
Counterbalancing |
|
|
Anticipation Effects |
In an experimental design counterbalancing (like ABBA) a participant may notice a pattern and change their responses accordingly. |
|
|
In an experimental design counterbalancing (like ABBA) a participant may notice a pattern and change their responses accordingly.
|
Anticipation Effects |
|
|
Types of Complete Design Counterbalancing |
Block Randomization and ABBA Counterbalancing |
|
|
Block Randomization Counterbalancing |
Each block contains all the conditions in a random order that continues until all stimuli are administered |
|
|
Counterbalancing where each block contains all the conditions in a random order that continues until all stimuli are administered |
Block Randomization Counterbalancing |
|
|
ABBA Counterbalancing |
Conditions in one sequence and then in reverse |
|
|
Counterbalancing where a person does all conditions first in one order and then in exactly reverse order. |
ABBA Counterbalancing |
|
|
Is ABBA Counterbalancing complete or incomplete? |
Complete |
|
|
Is Block Randomization complete or incomplete counterbalancing? |
Complete |
|
|
Is All Possible Orders Counterbalancing complete or incomplete? |
incomplete |
|
|
Is Random Starting Order with Rotation a complete or incomplete form of counterbalancing? |
incomplete |
|
|
What is the difference between Complete and Incomplete Counterbalancing |
In complete counterbalancing the practice effects are balanced for *each* individual, but in incomplete counterbalancing the practice effects are balanced for the group. |
|
|
How do you find the number of participants necessary for an all possible orders method of counterbalancing? |
N! |
|
|
How do you test for differential transfer |
by comparing results of repeated measures and independent groups
|
|
|
Confidence interval
|
indicates the range of values which we can expect to contain a population value within a specified degree of confidence.
|
|
|
indicates the range of values which we can expect to contain a population value within a specified degree of confidence.
|
Confidence interval
|
|
|
standard deviation
|
a measure of dispersion that indicates approx. how far on average scores deviate from the mean
|
|
|
a measure of dispersion that indicates approx. how far on average scores deviate from the mean
|
standard deviation
|
|
|
what is the most common measure of dispersion?
|
standard deviation
|
|
|
the score that appears most frequently in the distribution
|
mode
|
|
|
mode
|
the score that appears most frequently in the distribution
|
|
|
median
|
the midpoint of a distribution, above which half the scores fall, with the other half falling below.
|
|
|
the arithmetic average
|
mean
|
|
|
mean
|
the arithmetic average
|
|
|
what is the most commonly used measure of central tendency?
|
mean
|
|
|
Type 2 error
|
the probability of failing to reject the H_0 when it is false
|
|
|
the probability of failing to reject the H_0 when it is false
|
Type 2 error
|
|
|
Informed consent is an explicitly expressed willingness to participate in research based on these factors;
|
clear understanding of the nature of the researchconsequences of not participatingall factors that might influence willingness to participate
|
|
|
what makes it possible to predict values of one variable if you know the values of the second variable?
|
correlation
|
|
|
exists when 2 measures of the same people, events, or things vary together
|
correlation
|
|
|
independent group design
|
each group in the experiment represents a different condition as defined by the level of the IV.RandomMatchedNatural
|
|
|
each group in the experiment represents a different condition as defined by the level of the IV.
|
independent group design
|
|
|
3 types of independent group design
|
Randommatchednatural
|
|
|
this design can be random, matched, or natural.
|
independent group design
|
|
|
Relationship between effect size and sample size
|
the effect size is the strength of the relationship between the IV and the DV THAT IS INDEPENDENT OF SAMPLE SIZE.
|
|
|
effect size
|
index of the strength of the relationship between the IV and the DV that is independent of sample size.
|
|
|
index of the strength of the relationship between the IV and the DV that is independent of sample size.
|
effect size
|
|
|
this error is equal to the level of significance or alpha
|
Type 1 Error
|
|
|
Type 1 error is equal to
|
the level of significance or alpha
|
|
|
Type 1 Error
|
the probability of REJECTING the H_0 when it is TRUE. = to the level of significance or alpha.
|
|
|
the probability of REJECTING the H_0 when it is TRUE.
|
Type 1 Error
|
|
|
What are the 3 things that impact power in a study?
|
Size of the Treatment effectSample sizeAlpha (level of significance selected)(S.A.SS. is what affects power)
|
|
|
Null Hypothesis
|
(H_0) = assumption used as the first step in statistical inference whereby the IV is said to have no effect.
|
|
|
Error Variation
|
Chance
|
|
|
The probability when testing the null hypothesis that is used to indicate whether an outcome is statistically significant.
|
Level of Significance
|
|
|
Level of Significance
|
The probability when testing the null hypothesis that is used to indicate whether an outcome is statistically significant.
|
|
|
alpha =
|
level of significance
|
|
|
level of significance =
|
alpha
|
|
|
The probability of rejecting the null hypothesis when it is true, equal to the level of significance, or alpha.
|
Type I Error
|
|
|
Statistically Significant
|
When the probability of an obtained difference in an experiment is smaller than would be expected if error variation alone were assumed to be responsible for the difference, the difference is statistically significant.
|
|
|
Type II Error
|
The probability of failing to reject the null hypothesis when it is false.
|
|
|
The probability of failing to reject the null hypothesis when it is false.
|
Type II Error
|
|
|
the "truthfulness" of a measure; one that measures what it claims to measure.
|
validity
|
|
|
validity
|
the "truthfulness" of a measure; one that measures what it claims to measure.
|
|
|
Threats to Internal Validity
|
Possible causes that must be controlled so a clear cause-effect inference can be made.
|
|
|
Possible causes that must be controlled so a clear cause-effect inference can be made.
|
Threats to Internal Validity
|
|
|
Reversal Design =
|
ABAB Design =
|
|
|
ABAB Design =
|
Reversal Design =
|
|
|
when a measurement is consistent
|
Reliability
|
|
|
Reliability
|
when a measurement is consistent
|
|
|
Changes that participants undergo with repeated testing. They are the summation of both positive ( ex - familiarity with a test) and negative (ex - boredom) factors associated with repeated measures.
|
Practice Effects
|
|
|
Practice Effects
|
Changes that participants undergo with repeated testing. They are the summation of both positive ( ex - familiarity with a test) and negative (ex - boredom) factors associated with repeated measures.
|
|
|
Probability in a statistical test that a false null hypothesis will be rejected; it is related to:Sample SizeEffect SizeAlpha Level
|
power
|
|
|
ABAB Design
|
(AKA - REVERSAL DESIGN)A single-subject experimental design where an initial baseline stage (A) is followed by a treatment stage (B); the researcher observes whether behavior changes on introduction of the treatment, reverses when the treatment is withdrawn, and improves again when the treatment is reintroduced.
|
|
|
(AKA - REVERSAL DESIGN)A single-subject experimental design where an initial baseline stage (A) is followed by a treatment stage (B); the researcher observes whether behavior changes on introduction of the treatment, reverses when the treatment is withdrawn, and improves again when the treatment is reintroduced.
|
ABAB Design
|
|
|
Alternative Hypothesis =
|
H_1
|
|
|
The hypothesis that states a treatment *does* have an effect.
|
Alternative Hypothesis
|
|
|
The probability of getting a particular effect if the H_0 were true.
|
P-value
|
|
|
p-value
|
The probability of getting a particular effect if the H_0 were true.
|
|
|
_______ Is based on probability.
|
NHST (This means it can support a hypothesis, but never prove or disprove one; there is always a possibility for error)
|
|
|
What does it mean to reject the null hypothesis?
|
If you reject the null hypothesis it means that you have observed a sample that disagrees with the null hypothesis enough to allow to you to conclude it is false and the alternate hypothesis is true
|
|
|
Cohen's d
|
THE EFFECT SIZE OF THE DIFFERENCE BETWEEN GROUPS.
|
|
|
THE EFFECT SIZE OF THE DIFFERENCE BETWEEN GROUPS.
|
Cohen's d
|
|
|
Dependent Variable
|
A measure of behavior used to assess the effect of the independent variable.
|
|
|
A measure of behavior used to assess the effect of the independent variable.
|
Dependent Variable
|
|
|
Independent Variable
|
Factor for which the researcher manipulates at least two levels in order to determine its effect on behavior
|
|
|
MEASURES THE EFFECT SIZE (STRENGTH) OF A CORRELATION.
|
Pearson's r
|
|
|
Pearson's r
|
MEASURES THE EFFECT SIZE (STRENGTH) OF A CORRELATION.
|
|
|
More ____________ makes it more likely you will detect an effect.
|
power
|
|
|
Control requires _____________
|
balanced samples
|
|
|
What do you need in order to have a causal inference?
|
Co-variationTime-Order Relationship Elimination of Alternative Explanations
|
|
|
Time-Order Relationship
|
IV before DV
|
|
|
IV before DV
|
Time-Order Relationship
|
|
|
Co-variation
|
Performance on the DV changes for different levels of the IV (ex - exam score differs for those who had caffeine compared to those who do not.)
|
|
|
Performance on the DV changes for different levels of the IV (ex - exam score differs for those who had caffeine compared to those who do not.)
|
Co-variation
|
|
|
How do we eliminated plausible alternative explanations for a causal inference?
|
Elimination of plausible alternative explanations is only achieved through balanced groups.
|
|
|
Why are experiments the best way to determine causality?
|
Because if control is sufficient and all causal requirements are met, then any differences between the levels of the IV must be caused by the independent variable.
|
|
|
A study that is internally valid is free from ________
|
confounds
|
|
|
Attrition is a threat to _______
|
internal validity
|
|
|
Occurs when subjects are lost differentially across the conditions of the experiment as the result of some characteristic of each subject that is related to the outcome of the study.
|
Selective Subject Loss
|
|
|
External Validity
|
The extent to which the results of a study can be generalized to different populations, settings, and conditions.
|
|
|
Matched-Groups Design
|
Type of Independent Groups Design in which the researcher forms comparable groups by matching subjects on a pretest task and then randomly assigning the members of those matched sets of subjects to the conditions of the experiment.
|
|
|
Type of independent groups design in which the conditions represent the selected levels of a naturally occurring independent variable, for example, the individual differences variable age.
|
Natural Groups Deign
|
|
|
Repeated Measures Design
|
Research designs in which each subject participates in all conditions of the experiment (ex - all measurement is repeated on the same subject.)
|
|
|
Control technique for distributing (balancing) practice effects across the conditions of a repreated measures design. How it is accomplished depends on whether or not the study is a complete or incomplete repeated measures design.
|
Counterbalancing
|
|
|
Counterbalancing
|
Control technique for distributing (balancing) practice effects across the conditions of a repreated measures design. How it is accomplished depends on whether or not the study is a complete or incomplete repeated measures design.
|
|
|
How do you figure out how many orders would be required to counterbalance with an All Possible Orders form of Counterbalancing?
|
N!
|
|
|
N!
|
How do you figure out how many orders would be required to counterbalance with an All Possible Orders form of Counterbalancing?
|
|
|
Potential problem in repeated measures designs when performance in one condition differs depending on which of two other conditions preceeds it.
|
Differential Transfer
|
|
|
Differential Transfer
|
Potential problem in repeated measures designs when performance in one condition differs depending on which of two other conditions preceeds it.
|
|
|
THREAT TO THE REPRESENTATIVENESS OF A SAMPLE THAT OCCURS WHEN THE PROCEDURES USED TO SELECT A SAMPLE RESULT IN THE OVER OR UNDER REPRESENTATION OF A SIGNIFICANT SEGMENT OF THE POPULATION.
|
SELECTION BIAS
|
|
|
SELECTION BIAS
|
THREAT TO THE REPRESENTATIVENESS OF A SAMPLE THAT OCCURS WHEN THE PROCEDURES USED TO SELECT A SAMPLE RESULT IN THE OVER OR UNDER REPRESENTATION OF A SIGNIFICANT SEGMENT OF THE POPULATION.
|
|
|
SET OF ALL CASES OF INTEREST
|
POPULATION
|
|
|
POPULATION
|
SET OF ALL CASES OF INTEREST
|
|
|
ANOVA |
THE ANALYSIS OF VARIANCE IS THE MOST COMMONLY USED INFERENTIAL TEST OF EXAMINING A H_0 WHEN COMPARING MORE THAN 2 MEANS IN A SINGLE FACTOR STUDY, OR IN STUDIES WITH MORE THAN ONE IV. THE ANOVA IS BASED ON ANALYZING DIFFERENT SOURCES OF VARIATION IN AN EXPERIMENT |
|
|
THE MOST COMMONLY USED INFERENTIAL TEST OF EXAMINING A H_0
|
ANOVA
|
|
|
BASELINE STAGE |
FIRST STAGE OF A SINGLE SUBJECT EXPERIMENT IN WHICH A RECORD IS MADE OF THE PERSON'S BEHAVIOR PRIOR TO ANY INTERVENTION |
|
|
FIRST STAGE OF A SINGLE SUBJECT EXPERIMENT IN WHICH A RECORD IS MADE OF THE PERSON'S BEHAVIOR PRIOR TO ANY INTERVENTION
|
BASELINE STAGE
|
|
|
CEILING AND FLOOR EFFECT |
WHEN A RESEARCHER CANNOT MEASURE THE EFFETCTS OF AN IV OR INTERACTION BECAUSE PERFORMANCE HAS HIT A MAX OR MINIMUM |
|
|
WHEN A RESEARCHER CANNOT MEASURE THE EFFETCTS OF AN IV OR INTERACTION BECAUSE PERFORMANCE HAS HIT A MAX OR MINIMUM
|
CEILING AND FLOOR EFFECT
|
|
|
CONFIDENCE INTERVAL |
THE RANGE OF VALUES WE CAN EXPECT TO CONTAIN A POPULATION VALUE WITHIN A SPECIFIED DEGREE OF CONFIDENCE |
|
|
THE RANGE OF VALUES WE CAN EXPECT TO CONTAIN A POPULATION VALUE WITHIN A SPECIFIED DEGREE OF CONFIDENCE
|
CONFIDENCE INTERVAL
|
|
|
CONSTRUCT |
A CONCEPT OR IDEA USED IN PSYCH THEORIES TO EXPLAIN BEHAVIOR OR MENTAL PROCESSES; AGGRESSION, DEPRESSION, INTELLIGENCE, MEMORY, PERSONALITY |
|
|
A CONCEPT OR IDEA USED IN PSYCH THEORIES TO EXPLAIN BEHAVIOR OR MENTAL PROCESSES; AGGRESSION, DEPRESSION, INTELLIGENCE, MEMORY, PERSONALITY
|
CONSTRUCT
|
|
|
CONTAMINATION |
OCCURS WHEN THERE IS COMMUNICATION OF INFORMATION BETWEEN GROUPS OF PARTICIPANTS |
|
|
OCCURS WHEN THERE IS COMMUNICATION OF INFORMATION BETWEEN GROUPS OF PARTICIPANTS
|
CONTAMINATION
|
|
|
EXTERNAL VALIDITY |
THE EXTENT TO WHICH THE RESULTS OF A STUDY CAN BE GENERALIZED TO DIFFERENT POPULATIONS, SETTINGS, AND CONDITIONS |
|
|
THE EXTENT TO WHICH THE RESULTS OF A STUDY CAN BE GENERALIZED TO DIFFERENT POPULATIONS, SETTINGS, AND CONDITIONS
|
EXTERNAL VALIDITY
|
|
|
FACTORIAL DESIGN AKA |
COMPLEX DESIGN |
|
|
COMPLEX DESIGN AKA |
FACTORIAL DESIGN |
|
|
F-TEST |
IN ANOVA THE RATIO OF BETWEEN GROUP VARIANTION AND WITHIN GROUP OR ERROR VARIATION |
|
|
IN ANOVA THE RATIO OF BETWEEN GROUP VARIANTION AND WITHIN GROUP OR ERROR VARIATION
|
F-TEST
|
|
|
HISTORY |
THE OCCURRENCE OF AN EVENT OTHER THAN THE TREATMENT THAT CAN THREATEN INTERNAL VALIDITY IF IT PRODUCES CHANGES IN THE PARTICIPANT'S BEHAVIOR |
|
|
THE OCCURRENCE OF AN EVENT OTHER THAN THE TREATMENT THAT CAN THREATEN INTERNAL VALIDITY IF IT PRODUCES CHANGES IN THE PARTICIPANT'S BEHAVIOR
|
HISTORY
|
|
|
IDEOGRAPHIC APPROACH |
INTENSIVE STUDY OF AN INDIVIDUAL, WITH AN EMPHASIS ON BOTH INDIVIDUAL UNIQUENESS AND LAWFULNESS |
|
|
INTENSIVE STUDY OF AN INDIVIDUAL, WITH AN EMPHASIS ON BOTH INDIVIDUAL UNIQUENESS AND LAWFULNESS
|
IDEOGRAPHIC APPROACH
|
|
|
INDEPENDENT GROUPS DESIGN |
EACH GROUP IN THE EXPERIMENT REPRESENTS A DIFFERENT CONDITION AS DEFINED BY THE LEVEL OF THE IV |
|
|
EACH GROUP IN THE EXPERIMENT REPRESENTS A DIFFERENT CONDITION AS DEFINED BY THE LEVEL OF THE IV
|
INDEPENDENT GROUPS DESIGN
|
|
|
INSTRUMENTATION |
CHANGES OVER TIME CAN TAKE PLACE NOT ONLY IN PARTICIPANTS, BUT ALSO IN THE INSTRUMENTS. THESE CHANGES DUE TO INSTRUMENTATION CAN THREATEN INTERNAL VALIDITY IF THEY CANNOT BE SEPARATED FROM THE TREATMENT EFFECT |
|
|
CHANGES OVER TIME CAN TAKE PLACE NOT ONLY IN PARTICIPANTS, BUT ALSO IN THE INSTRUMENTS
|
INSTRUMENTATION
|
|
|
INTERACTION EFFECT |
WHEN THE EFFECT OF ONE IV DIFFERS DEPENDING ON THE LEVEL OF A SECOND IV |
|
|
WHEN THE EFFECT OF ONE IV DIFFERS DEPENDING ON THE LEVEL OF A SECOND IV
|
INTERACTION EFFECT
|
|
|
MAIN EFFECT |
OVERALL EFFECT OF AN IV IN A COMPLEX DESIGN |
|
|
OVERALL EFFECT OF AN IV IN A COMPLEX DESIGN
|
MAIN EFFECT
|
|
|
TYPE OF INDEPENDENT GROUPS DESIGN IN WHICH THE RESEARCHER FORMS COMPARABLE GROUPS BY MATCHING SUBJECTS ON A PRETEST TASK AND THEN RANDOMLY ASSIGNING THE MEMBERS OF THESE SETS TO THE CONDITIONS OF THE EXPERIMENTS
|
MATCHED GROUPS DESIGN
|
|
|
MATURATION |
CHANGES ASSOCIATED WITH THE PASSAGE OF TIME. CHANGES PARTICIPANTS UNDERGO IN AN EXPERIMENT THAT ARE DUE TO MATURATION AND NOT DUE TO TREATMENT CAN THREATEN INTERNAL VALIDITY |
|
|
CHANGES ASSOCIATED WITH THE PASSAGE OF TIME.
|
MATURATION
|
|
|
MEASURES OF CENTRAL TENDENCY |
MEAN, MEDIAN, MODE |
|
|
MEDIAN |
THE MIDDLE POINT IN A DISTRIBUTION WHERE HALF FALL ABOVE AND HALF OF THE SCORES FALL BELOW |
|
|
THE MIDDLE POINT IN A DISTRIBUTION WHERE HALF FALL ABOVE AND HALF OF THE SCORES FALL BELOW
|
MEDIAN
|
|
|
MODE |
APPEARS MOST FREQUENTLY |
|
|
SCORE THAT APPEARS MOST FREQUENTLY |
MODE |
|
|
MULTIMETHOD APPROACH |
APPROACH TO HYPOTHESIS TESTING THAT SEEKS EVIDENCE BY COLLECTING DATA USING SEVERAL DIFFERENT RESEARCH PROCEDURES AND MEASURES OF BEHAVIOR; A RECOGNITION OF THE FACT THAT ANY SINGLE OBSERVATION OF BEHAVIOR CAN RESULT FROM SOME ARTIFACT OF THE MEASURING PROCESS |
|
|
APPROACH TO HYPOTHESIS TESTING THAT SEEKS EVIDENCE BY COLLECTING DATA USING SEVERAL DIFFERENT RESEARCH PROCEDURES AND MEASURES OF BEHAVIOR; A RECOGNITION OF THE FACT THAT ANY SINGLE OBSERVATION OF BEHAVIOR CAN RESULT FROM SOME ARTIFACT OF THE MEASURING PROCESS
|
MULTIMETHOD APPROACH
|
|
|
NATURAL GROUPS DESIGN |
INDEPENDENT GROUPS DESIGN IN WHICH THE CONDITIONS REPRESENT THE SELECTED LEVELS OF A NATURALLY OCCURING IV -- AGE, RACE, ETC |
|
|
INDEPENDENT GROUPS DESIGN IN WHICH THE CONDITIONS REPRESENT THE SELECTED LEVELS OF A NATURALLY OCCURING IV -- AGE, RACE, ETC
|
NATURAL GROUPS DESIGN
|
|
|
NOMOTHETIC APPROACH |
APPROACH TO RESEARCH THAT SEEKS TO ESTABLISH BROAD GENERALIZATIONS OR LAWS THAT APPLY TO LARGE GROUPS. THE AVERAGE OR TYPICAL PERFORMANCE IS EMPHASIZED |
|
|
APPROACH TO RESEARCH THAT SEEKS TO ESTABLISH BROAD GENERALIZATIONS OR LAWS THAT APPLY TO LARGE GROUPS. THE AVERAGE OR TYPICAL PERFORMANCE IS EMPHASIZED
|
NOMOTHETIC APPROACH
|
|
|
NONEQUIVALENT CONTROL GROUP DESIGN |
QUASI PROCEDURE IN WHICH A COMPARISON IS MADE BETWEEN CONTROL AND TREATMENT GROUPS THAT HAVE BEEN ESTABLISHED ON SOME BASIS OTHER THAN RANDOM ASSIGNMENT |
|
|
QUASI PROCEDURE IN WHICH A COMPARISON IS MADE BETWEEN CONTROL AND TREATMENT GROUPS THAT HAVE BEEN ESTABLISHED ON SOME BASIS OTHER THAN RANDOM ASSIGNMENT
|
NONEQUIVALENT CONTROL GROUP DESIGN
|
|
|
OMNIBUS F-TEST |
THE INITIAL OVERALL ANALYSIS BASED ON ANOVA |
|
|
THE INITIAL OVERALL ANALYSIS BASED ON ANOVA
|
OMNIBUS F-TEST
|
|
|
OPERATIONAL DEFINITION |
PROCEDURE WHEREBY A CONCEPT IS DEFINED SOLELY IN TERMS OF THE OBSERVABLE PROCEDURES USED TO PRODUCE AND MEASURE IT. |
|
|
PROCEDURE WHEREBY A CONCEPT IS DEFINED SOLELY IN TERMS OF THE OBSERVABLE PROCEDURES USED TO PRODUCE AND MEASURE IT.
|
OPERATIONAL DEFINITION
|
|
|
SOURCE OF EVIDENCE THAT IS BASED ON THE REMNANTS, FRAGMENTS, AND PRODUCTS OF PAST BEHAVIOR
|
PHYSICAL TRACES
|
|
|
QUASI EXPERIMENTS |
PROCEDURES THAT LOOK LIKE TRUE EXPERIMENTS BUT ARE LACKING IN THE DEGREE OF CONTROL |
|
|
PROCEDURES THAT LOOK LIKE TRUE EXPERIMENTS BUT ARE LACKING IN THE DEGREE OF CONTROL
|
QUASI EXPERIMENTS
|
|
|
RANDOM GROUPS DESIGN |
MOST COMMON TYPE OF INDEPENDENT GROUPS DESIGN WHERE SUBJECTS ARE RANDOMLY ASSIGNED TO EACH GROUP SUCH THAT GROUPS ARE CONSIDERED COMPARABLE AT THE START OF THE STUDY |
|
|
MOST COMMON TYPE OF INDEPENDENT GROUPS DESIGN WHERE SUBJECTS ARE RANDOMLY ASSIGNED TO EACH GROUP SUCH THAT GROUPS ARE CONSIDERED COMPARABLE AT THE START OF THE STUDY
|
RANDOM GROUPS DESIGN
|
|
|
RANGE = |
HIGHEST - LOWEST |
|
|
RELEVANT INDEPENDENT VARIABLE |
IV THAT HAS BEEN SHOWN TO INFLUENCE BEHAVIOR, EITHER DIRECTLY THROUGH A MAIN EFFECT, OR INDIRECTLY THROUGH INTERACTION WITH A SECOND IV |
|
|
IV THAT HAS BEEN SHOWN TO INFLUENCE BEHAVIOR, EITHER DIRECTLY THROUGH A MAIN EFFECT, OR INDIRECTLY THROUGH INTERACTION WITH A SECOND IV
|
RELEVANT INDEPENDENT VARIABLE
|
|
|
RELIABILITY |
CONSISTENT |
|
|
CONSISTENT
|
RELIABILITY
|
|
|
REPEATED MEASURES DESIGNS |
EACH SUBJECT PARTICIPATES IN ALL CONDITIONS |
|
|
EACH SUBJECT PARTICIPATES IN ALL CONDITIONS
|
REPEATED MEASURES DESIGNS
|
|
|
REPEATED MEASURES (WITHIN SUBJECTS) T TEST |
INFERENTIAL TEST FOR COMPARING TWO MEANS FROM THE SAME GROUP OF SUBJECT OR TWO DIFFERENT MATCHED GROUPS ON SOME MEASURE RELATED TO THE DV |
|
|
INFERENTIAL TEST FOR COMPARING TWO MEANS FROM THE SAME GROUP OF SUBJECT OR TWO DIFFERENT MATCHED GROUPS ON SOME MEASURE RELATED TO THE DV
|
REPEATED MEASURES (WITHIN SUBJECTS) T TEST
|
|
|
SCIENTIFIC METHOD |
APPROACH TO KNOWLEDGE THAT EMPHASIZES EMPIRICAL RATHER THAN INTUITIVE PROCESSES, TESTABLE HYPOTHESIS, SYSTEMATIC AND CONTROLLED OBSERVATION OF OPERATIONALLY DEFINED PHENOMENA, VALID AND RELIABLE MEASURES, OBJECTIVE AND SKEPTICAL SCIENTISTS |
|
|
APPROACH TO KNOWLEDGE THAT EMPHASIZES EMPIRICAL RATHER THAN INTUITIVE PROCESSES, TESTABLE HYPOTHESIS, SYSTEMATIC AND CONTROLLED OBSERVATION OF OPERATIONALLY DEFINED PHENOMENA, VALID AND RELIABLE MEASURES, OBJECTIVE AND SKEPTICAL SCIENTISTS
|
SCIENTIFIC METHOD
|
|
|
SELECTION |
A THREAT TO INTERNAL VALIDITY WHERE DIFFERENCES EXIST BETWEEN THE KINDS OF PPL IN ONE GROUP AND ANOTHER AT THE START OF THE STUDY |
|
|
A THREAT TO INTERNAL VALIDITY WHERE DIFFERENCES EXIST BETWEEN THE KINDS OF PPL IN ONE GROUP AND ANOTHER AT THE START OF THE STUDY
|
SELECTION
|
|
|
SIMPLE INTERRUPTED TIME SERIES DESIGN |
QUASI IN WHICH CHANGES IN A DV ARE OBSERVED FOR A PERIOD OF TIME BOTH BEFORE AND AFTER TX |
|
|
QUASI IN WHICH CHANGES IN A DV ARE OBSERVED FOR A PERIOD OF TIME BOTH BEFORE AND AFTER TX
|
SIMPLE INTERRUPTED TIME SERIES DESIGN
|
|
|
SIMPLE MAIN EFFECT |
EFFECT OF ONE IV AT ONE LEVEL OF A SECOND IV IN A COMPLEX DESIGN |
|
|
EFFECT OF ONE IV AT ONE LEVEL OF A SECOND IV IN A COMPLEX DESIGN
|
SIMPLE MAIN EFFECT
|
|
|
SINGLE FACTOR INDEPENDENT GROUPS DESIGN |
EXPERIMENT THAT INVOLVES INDEPENDENT GROUPS WITH ONE IV |
|
|
EXPERIMENT THAT INVOLVES INDEPENDENT GROUPS WITH ONE IV
|
SINGLE FACTOR INDEPENDENT GROUPS DESIGN
|
|
|
SINGLE SUBJECT EXPERIMENT |
PROCEDURE THAT FOCUSES ON BEHAVIOR CHANGE IN ONE PERSON BY CONTRASTING CONDITIONS WITHIN THAT PERSON WHILE CONTINUOUSLY MONITORING BEHAVIOR |
|
|
PROCEDURE THAT FOCUSES ON BEHAVIOR CHANGE IN ONE PERSON BY CONTRASTING CONDITIONS WITHIN THAT PERSON WHILE CONTINUOUSLY MONITORING BEHAVIOR
|
SINGLE SUBJECT EXPERIMENT
|
|
|
SITUATION SAMPLING |
RANDOM OR SYSTEMATIC SELECTION OF SITUATIONS WHERE OBSERVATIONS ARE MADE WITH THE GOAL OF REPRESENTATIVENESS ACROSS CIRCIMSTANCES, LOCATIONS, AND CONDITIONS |
|
|
RANDOM OR SYSTEMATIC SELECTION OF SITUATIONS WHERE OBSERVATIONS ARE MADE WITH THE GOAL OF REPRESENTATIVENESS ACROSS CIRCIMSTANCES, LOCATIONS, AND CONDITIONS
|
SITUATION SAMPLING
|
|
|
SMALL N RESEARCH = |
SINGLE SUBJECT EXPERIMENT |
|
|
SINGLE SUBJECT EXPERIMENT =
|
SMALL N RESEARCH =
|
|
|
TESTING |
TAKING A TEST HAS AN EFFECT ON SUBSEQUENT TESTING. THE TESTING EFFECT CAN THREATEN INTERNAL VALIDITY IF IT CANNOT BE SEPARATED FROM THE TX |
|
|
TAKING A TEST HAS AN EFFECT ON SUBSEQUENT TESTING.
|
TESTING
|
|
|
POSSIBLE CAUSES THAT MUST BE CONTROLLED IN ORDER TO INFER CAUSALITY
|
THREATS TO INTERNAL VALIDTY
|
|
|
TIME SERIES WITH NONEQUIVALENT GROUPS DESIGN |
QUASI THAT IMPROVES ON THE VALIDITY OF A SIMPLE TIME SERIES DESIGN BY INCLUDING A NONEQUIVALENT CONTROL GROUP WHERE BOTH GROUPS ARE OBSERVED BEFORE AND AFTER TX |
|
|
QUASI THAT IMPROVES ON THE VALIDITY OF A SIMPLE TIME SERIES DESIGN BY INCLUDING A NONEQUIVALENT CONTROL GROUP WHERE BOTH GROUPS ARE OBSERVED BEFORE AND AFTER TX
|
TIME SERIES WITH NONEQUIVALENT GROUPS DESIGN
|
|
|
INCOMPLETE REPEATED MEASURES DESIGNS |
|
|
|
COMPLETE REPEATED MEASURES DESIGNS |
ABBA AND BLOCK RANDOMIZATION |
|
|
TYPES OF INDEPENDENT GROUPS STUDIES |
|
|
|
MAIN EFFECT |
AN EFFECT OF A SINGLE IV ALONE |
|
|
AN EFFECT OF A SINGLE IV ALONE
|
MAIN EFFECT
|
|
|
INTERACTION |
WHEN THE EFFECT OF AN IV IS DIFFERENT AT DIFFERENT LEVELS OF AN IV. (IT DEPENDS) |
|
|
WHEN THE EFFECT OF AN IV IS DIFFERENT AT DIFFERENT LEVELS OF AN IV. (IT DEPENDS)
|
INTERACTION
|
|
|
REQUIREMENTS OF THE SINGLE SUBJECT EXPERIMENT |
|
|
|
8 THREATS TO INTERNAL VALIDITY |
A. R.I.M.S.H.O.T. IS A THREAT TO INTERNAL VALIDITY |
|
|
TYPES OF QUASI DESIGNS |
|
|
|
PURPOSE OF THE RESEARCH REPORT |
|
|
|
STRUCTURE OF A RESEARCH REPORT |
|
A.I.M RESULTS DISCUSS - REF NO TAB |
|
WHAT THREE CATEGORIES GO UNDER METHODS |
PARTICIPANTS, MEASURES, AND PROCEDURES |
|
|
PARTICIPANTS, MEASURES, AND PROCEDURES GO UNDER WHAT HEADING
|
METHODS |
|
|
WHAT GOES UNDER THE DISCUSSION HEADING |
|
|
|
LIMITATIONS PLUS IMPLICATIONS AND CONCLUSIONS GO UNDER WHAT HEADING |
DISCUSSION |
|
|
3 WAYS TO COMMUNICATE RESEARCH TO OTHERS |
|
|
|
WHAT IS THE PURPOSE OF THE ABSTRACT |
SUMMARIZES THE ENTIRE REPORT
|
|
|
PARTS OF THE ABSTRACT |
|
|
|
WHAT PART OF RESEARCH DO YOU WRITE LAST |
ABSTRACT |
|
|
INFO IN THE INTRO |
|
|
|
METHODS |
DETAILS OF HOW THE STUDY WAS CONDUCTED IN ORDER TO ANSWER THE QUESTIONS IN THE INTRO |
|
|
DETAILS OF HOW THE STUDY WAS CONDUCTED IN ORDER TO ANSWER THE QUESTIONS IN THE INTRO |
METHODS |
|
|
WHAT GOES UNDER THE METHODS HEADING? |
|
|
|
RESULTS |
DESCRIBES WHAT WAS FOUND USING THE DATA GATHERED THROUGH PROCEDURES OUTLINED IN METHODS |
|
|
DESCRIBES WHAT WAS FOUND USING THE DATA GATHERED THROUGH PROCEDURES OUTLINED IN METHODS
|
RESULTS
|
|
|
RESULTS INCLUDE |
|
|
|
TABLES AND FIGURES HEADING INCLUDES |
|
|
|
IN THE DISCUSSION HEADING YOU SHOULD AVOID |
CAUSAL STATEMENTS AND EXCESSIVE THEORIZING |
|
|
IN THE DISCUSSION HEADING YOU SHOULD |
SUMMARIZE THE RESULTS BRIEFLY AND DISCUSS THE MEANING/INTERPRETATION OF THE RESULTS |
|
|
LIMITATIONS |
HOW THE RESULTS CAN BE GENERALIZED AND HOW THEY CANNOT AND HOW FUTURE RESEARCH MIGHT CORRECT THESE LIMITATIONS |
|
|
CONCLUSIONS AND IMPLICATIONS |
EMPHASIZES THE BROAD IMPORTANCE AND SUGGESTS APPLICATIONS AND IMPLICATIONS FOR FUTURE RESEARCH |
|
|
INTRO |
EXPLAINS THE PURPOSE AND RATIONALE OF THE STUDY BASED ON PREVIOUS LITERATURE AND OUTLINES HYPOTHESES TO BE TESTED |
|
|
EXPLAINS THE PURPOSE AND RATIONALE OF THE STUDY BASED ON PREVIOUS LITERATURE AND OUTLINES HYPOTHESES TO BE TESTED
|
INTRO
|
|
|
DISCUSSION |
INTERPRETATION OF THE RESULTS AND RELATES THEM TO CURRENT KNOWLEDGE AND REAL WORLD ISSUES |
|
|
INTERPRETATION OF THE RESULTS AND RELATES THEM TO CURRENT KNOWLEDGE AND REAL WORLD ISSUES
|
DISCUSSION
|
|
|
REFERENCES |
LISTS ALL RESEARCH OR WORK MENTIONED IN ANY SECTION |
|
|
LISTS ALL RESEARCH OR WORK MENTIONED IN ANY SECTION
|
REFERENCES
|
|
|
BASIC FORMATTING OF APA |
|
|
|
3 GENERAL GUIDELINES OF APA |
|
|
|
WHAT DISTINGUISHES A QUASI FROM TRUE EXPERIMENT |
LACK OF FULL CONTROL |
|
|
ALTERNATIVES TO NO TREATMENT CONTROL GROUP |
|
|
|
ALTERNATE ORDER OF TREATMENTS |
1 GROUP GETS TX1, THEN TX 2 AND THEN OTHER GROUP GETS TX 2 AND THEN TX 1 |
|
|
1 GROUP GETS TX1, THEN TX 2 AND THEN OTHER GROUP GETS TX 2 AND THEN TX 1 |
ALTERNATE ORDER OF TREATMENTS
|
|
|
WAIT LIST CONTROL |
ONE GROUPS GET TX AND IF IT IS BENEFICIAL THE CONTROL GROUP GETS THE TX LATER |
|
|
ONE GROUPS GET TX AND IF IT IS BENEFICIAL THE CONTROL GROUP GETS THE TX LATER
|
WAIT LIST CONTROL
|
|
|
SOLUTION TO HISTORY CONFOUND |
CONTROL GROUP |
|
|
SOLUTION TO MATURATION CONFOUND |
CONTROL GROUP |
|
|
SOLUTION TO TESTING CONFOUND |
|
|
|
SOLUTIONS TO INSTRUMENTATION CONFOUND |
ENSURE RELIABILITY AND VALIDITY OF INSTRUMENTS |
|
|
SOLUTION TO ATTRITION |
CAREFUL FOLLOW UP PROCEDURES AND COMPARE THOSE WHO DROPPED TO WHO REMAINS |
|
|
SOLUTION TO SELECTION CONFOUND |
RANDOMIZATION, BUT IF NOT POSSIBLE MATCHING GROUPS AND AWARENESS |
|
|
ADDITIVE EFFECTS WITH SELECTION |
WHEN ANY OF THE FIRST 6 THREATS TO INTERNAL VALIDITY EXIST FOR ONE GROUP BUT NOT THE OTHER |
|
|
WHEN ANY OF THE FIRST 6 THREATS TO INTERNAL VALIDITY EXIST FOR ONE GROUP BUT NOT THE OTHER
|
ADDITIVE EFFECTS WITH SELECTION
|
|
|
WHAT IS THE NUMBER ONE CONFOUND IN QUASI? |
ADDITIVE EFFECTS WITH SELECTION |
|
|
TWO TYPES OF CONTAMINATION |
|
|
|
DIFFUSION OF TREATMENT |
WHEN THE EXPERIMENTAL GROUP COMMUNICATES INFO TO THE CONTROL GROUP |
|
|
ADVANTAGES OF CASE STUDIES |
|
|
|
DISADVANTAGES OF CASE STUDIES |
|
|
|
A TESTIMONIAL IS A |
CASE STUDY |
|
|
SKINNERIAN ANALYSIS OF BEHAVIOR |
SMALL N DESIGNS |
|
|
SMALL N DESIGNS MAKE IT POSSIBLE TO ESTABLISH CAUSAL INFERENCES FOR |
ONLY THAT INDIVIDUAL OR SMALL GROUP |
|
|
HOW DO YOU MAKE A CAUSAL INFERENCE IN ABAB DESIGNS? |
|
|
|
WHEN DO WE USE MULTIPLE BASELINE DESIGNS? |
IN CASES WHERE REVERAL DOES NOT OCCUR OR WHEN IT WOULD BE UNETHICAL |
|
|
MULTIPLE BASELINES WON'T WORK IF... |
AN INTERVENTION GENERALIZES ACROSS INDIVIDUALS, BEHAVIORS, OR SITUATIONS |
|
|
ADVANTAGES OF A SINGLE SUBJECT EXPERIMENT |
|
|
|
DISADVANTAGES OF SINGLE SUBJECT |
|
|
|
THE PRESENCE OF NON PARALLEL LINES ON A GRAPH INDICATES THERE MAY BE |
AN INTERACTION BETWEEN VARIABLES |
|
|
THE KEY TO FINDING AN INTERACTION IS IF THE ANSWER YOU GET IS |
IT DEPENDS |
|
|
3 POSSIBLE EFFECTS IN COMPLEX DESIGNS |
|
|
|
INTERACTIONS = |
MODERATIONS |
|
|
2 LINES NOT OVERLAPPING IN A COMPLEX DESIGN GRAPH USUALLY MEANS |
A MAIN EFFECT |
|
|
SPURIOUS RELATIONSHIP |
WHAT EXISTS WHEN EVIDENCE FALSELY INDICATES THAT 2 OR MORE VARIABLES ARE ASSOCIATED |
|
|
WHAT EXISTS WHEN EVIDENCE FALSELY INDICATES THAT 2 OR MORE VARIABLES ARE ASSOCIATED
|
SPURIOUS RELATIONSHIP
|
|
|
IF THERE IS NO INTERACTION THE RESULTS MAY |
GENERALIZE TO ALL |
|
|
IF THERE IS AN INTERACTION EFFECT THE RESULTS |
ARE LIMITED TO ONE GROUP AND MAIN EFFECTS ARE LESS MEANINGFUL |
|
|
EACH IV CAN HAVE A |
MAIN EFFECT |
|
|
INTERACTIONS CAN REVEAL |
A HIDDEN EFFECT |
|
|
IF YOU SEE AN 'X' CONFIGURATION ON A LINE GRAPH IT USUALLY MEANS |
THERE ARE NO MAIN EFFECTS |
|
|
HAPPENS WHEN YOUR TEST IS TOO HARD |
FLOOR EFFECT |
|
|
CEILING AND FLOOR EFFECTS ARE DANGEROUS BECAUSE |
IT CAN MAKE IT LOOK LIKE THERE IS AN EFFECT WHEN THERE IS NOT |
|
|
CASE STUDIES ARE A ________ AND MAY NOT BE REPRESENTATIVE OF THE LARGER POPULATION |
SINGLE DATA POINT |
|
|
ARE SINGLE SUBJECT DESIGNS EXPERIMENTS? |
YES BECAUSE IT HAS BOTH MANIPULATION AND CONTROL |
|
|
TYPE OF EXPERIMENT OFTEN USED IN TREATMENT AND APPLIED SETTINGS |
QUASI |
|
|
SOLUTION TO NOVELTY EFFECTS |
|
|
|
NONEQUIVALENT CONTROL GROUP DESIGNS DON'T CONTROL FOR |
ADDITIVE EFFECTS WITH SELECTION |
|
|
REQUIRES AN ABRUPT DISCONTINUITY IN THE TIME SERIES |
INTERRUPTED TIME SERIES DESIGN |
|
|
DESIGN STYLE THAT IS THE BEST WAY TO JUDGE POLICY CHANGE |
INTERRUPTED TIME SERIES DESIGN WITH NONEQUIVALENT CONTROL GROUPS |
|
|
MIXED DESIGN |
1 DV AND 2 OR MORE IVs WHERE ONE OF THE IVs IS AN INDEPENDENT GROUPS DESIGN AND THE OTHER GROUP IS A REPEATED MEASURES DESIGN |
|
|
1 DV AND 2 OR MORE IVs WHERE ONE OF THE IVs IS AN INDEPENDENT GROUPS DESIGN AND THE OTHER GROUP IS A REPEATED MEASURES DESIGN
|
MIXED DESIGN
|
|
|
WHY SHOULD WE CARE ABOUT INTERACTIONS |
BECAUSE THEY PROVIDE INFO ABOUT LIMITS OF THE EFFECT OF AN IV. REAL EFFECTS CAN BE HIDDEN IF AN INTERACTION IS NOT ASSESSED AND INTERACTIONS HELP US UNDERSTAND GROUP DIFFERENCES |
|
|
WHEN IDENTIFYING MAIN EFFECTS YOU SHOULD CONFIRM WITH |
STATISTICS |
|
|
WHAT IS THE BEST WAY TO ILLUSTRATE MAIN AND INTERACTION EFFECTS? |
LINE GRAPH |
|
|
THIS GROUP DESIGN ANALYZES BETWEEN GROUP EFFECTS |
INDEPENDENT GROUP DESIGN |
|
|
COMPARES BETWEEN GROUP AND WITHIN GROUP VARIABILITY |
INDEPENDENT GROUP DESIGN |
|
|
IN A COMPLETE DESIGN PRACTICE EFFECTS ARE BALANCED OUT FOR |
EACH PARTICIPANT |
|
|
IN AN INCOMPLETE DESIGN PRACTICE EFFECTS ARE BALANCED OUT FOR |
THE GROUP |
|