Final Exam Mock Questions Flashcards
What is a frequency claim, how many variables does it involve and are the variables measured or manipulated?
A Frequency claim describes the particular rate or degree of a single variable
ONE measured variable
What is an association claim, how many variables does it involve and are the variables measured or manipulated?
A claim which argues that one level of a variable is correlated with a particular level of another variable
TWO measured variables (interval or ratio)
What is causal claim, how many variables does it involve and are the variables measured or manipulated?
A claim which argues that one variable is responsible for changing the other variable
TWO variables - one measure and one manipulated
What are the three criteria for a causal claim?
Temporal Precedence: The causal variable came before the affected variable
Covariance: Variables are associated
Internal Validity: There is no other explanation (confound)
What is construct validity?
An indication of how well the variables in a study were measured or manipulated (how well each variable was operationalized)
What is external validity?
An indication of how well the results of a study generalize to (or represent) the population
What is statistical validity?
An indication of how accurate and reliable the results of a study are (e.g. effect size and statistical significance)
What is internal validity?
An indication of how well an experiment avoided possible confounds
i.e. In a relationship between one variable (A) and another (B) the extent to which (A) rather than some other variable (C) is responsible for the change in (B)
What does it mean when the results of a study are statistical significant?
That it is unlikely that the results of the study were due to chance or come from a population where there is zero association
What is a Type I Error and what can be done to avoid it?
Concluding that there IS an effect in the population when there is NO effect
Lowering the Alpha Level reduces the chance of making a Type I Error
What is a Type II Error and what can be done to avoid it?
Concluding that there is NO effect in the populations when there IS an effect
Increasing the Alpha Level or Sample Size can help reduce the chance of making a Type II Error
What is the purpose of measuring Effect Size?
To describe the strength of the relationship between two variables (the size of difference between two groups)
What are the core principles of the TCPS?
Respect for Persons
Concern for Welfare
Justice
What does Concern for Welfare entail?
That researchers take precautions to protect the well-being of participants and that the risk and benefits of a study need to be weighed out (positive must outweigh negative)
What does Justice entail?
That researchers address issues of equity (participants must be selected equally, any reason to leave a particular group out must be justified) and those studied should benefit from what is learned
What does respect for Person entail?
That researchers receive informed consent from all participants and do not mislead (deception), coerce, or unduly influence anyone into participating in a study
What are the guidelines for animal care in experimental studies?
Replacement: where possible avoid using animal models
Reduction: minimize number of animals used
Refinement: minimize pain and distress
What does it mean when a researcher fabricates the data for their study?
They invented data to fit their hypothesis
What does it mean when a researcher falsifies the data for their study?
The researcher influenced the results of the study by removing data that doesn’t support their hypothesis or purposely influenced participants to respond/act a certain way
What is debriefing and when does it occur?
It is when you fully disclose the purpose of the study and everything that was conducted to participants after completion the study to address issues of withholding information, deception, and potential harmful effects
What is the difference between deception through omission and deception through commission?
With deception through omission you are withholding information from participants where as with deception through commission you are actively lying to participants
What is the difference between descriptive and inferential statistics?
Descriptive statistics is used to organize and summarize the properties of a data set from a sample while inferential statistics is used to take that same data and make predictions about the population the sample generalizes too
What computation describes how spread out scores from a sample are around the mean?
Variance (SD^2)
What computation describes how far, on average, scores are from the mean?
Standard Deviation (SD)
What is an ordinal scale?
A quantitative (numerical) scale with a ranked order in which it is unclear if the distances are equal
What is a ratio scale?
A quantitative (numerical scale) in which the intervals are equal and zero really means nothing Example: Weight
What is an interval scale?
A quantitative (numerical scale) in which the intervals are equal but there is no true zero Examples: 5-Star Hotel Rating or IQ
What does it mean for a measure to be reliable?
The results it provides are consistent - if you measure the same thing over and over you will get the same result over and over
What does it mean for a measure to be valid?
A measure is measuring the conceptual variable it is intended to measure
What does the correlation coefficient tell us?
The strength and direction of an association
What is the difference between probability and non-probability sampling?
In probability sampling everyone in the population has an equal chance of being selected for the sample and is random where as in non-probability sampling not everyone has a chance of being selected
What is stratified random sampling?
A probability sampling technique in which the population is divided into subgroups and random samples are taken from each subgroup
What is cluster sampling?
A probability sampling technique in which random clusters are chosen from the population and all participants in each cluster are used together as the sample
What is multistage sampling?
A probability sampling technique which chooses random clusters and then chooses participants at random from each cluster
What is purposive sampling?
A non-probability sampling technique which focuses only on choosing participants that meet predetermined criterion (take first people that meet criteria)
What is quota sampling?
A non-probability sampling technique in which the researcher identifies various subgroups and sets a target for each (e.g. race, gender, etc.)
What is convenience sampling?
a non-probability sampling technique in which participants are chosen based on who is most easily accessible
What is a selection effect?
When participants in one level of the independent variable are systematically different than those in the other level
If a researcher were to assign participants to levels by first pairing them up based on particular characteristics they share and then randomly assigning each person to different levels, what assignment method would they be using?
Matched-Groups Design
If a researcher were to use the same group for all levels of the independent variable, what assignment method would they be using?
Repeated Measure Design
If a researcher were to randomly divide the sample into each level of the independent variable, what assignment method would they be using?
Independent Groups Design
What are the possible negative effects of using a repeated measure that result from participants having to endure all levels of the independent variable?
Order Effect:
Practice Effect - participants improve with each level
Fatigue Effect - participants grow more and more tired and less interested with each level
Contrast Effect - Conditions from one level may carry over and interact with the next
What is counterbalancing?
Counterbalancing involves presenting participants to levels of the independent variable in different orders to prevent order effects
What is semantic differential scale?
A scale in which you rate a target object using a numeric scale with meaningful adjectives
What is a non-verbal scale?
A scale which uses images to represent the different options
What is Fence-Sitting and what are ways to avoid it?
When people play it safe by answering in the middle of the scale for all questions on a survey
You can avoid sitting by using an uneven amount of choices or forcing choice between only two options
What is yea-saying and what are ways to avoid it?
When someone selects yes (or agree) for every question
It can be avoided by using questions which reverse the order (by saying yes you now disagree)
What is Faking Good and what are ways to avoid it?
When someone answers questions on a survey based on what the deem to be socially acceptable
It can be avoided by making sure participants know their answers will be held anonymous
What are ways to avoid observer bias or observer effects?
Multiple, well-trained observers using reliable and clear rating scales (codebook)
Masked research design (observers are unaware of the purpose of the study or the hypothesis)
What is Reactivity and how can it be avoided?
When participants behaviour changes due to the presence of an observer
It can be avoided by: Blending in, waiting it out, or measuring the results of the behaviour instead of the behaviour itself
What are three ways to display correlational relationships?
Scatter Plots
Correlation Coefficient
Statistical Tests
What is a correlational design?
A study designed to determine if two quantitative variables are correlated
Ex. Does number of piercings correlated with final exam scores?
What is a Maturation Threat and what can be done to avoid it?
A threat to internal validity in which changes in behaviour emerge spontaneously over time as opposed to being an effect of the Independent Variable
Using a Comparison Group helps avoid Maturation Threats
What is a History Threat and what can be done to avoid it?
A threat to internal validity in which a historical or external event affects most members of the treatment group making it unclear whether the change was a result of the treatment of event
A comparison group can be used to avoid History Threats
What is a Regression Threat and what can be done to avoid it?
A threat to internal validity in which scores on a measure are extreme in one direction the first time but regress back to the mean (average) the next time
A comparison group can be used to help prevent Regression Threat
What is an Attrition Threat and what can be done to avoid it?
A threat to internal validity in which participants leave the study before it is over, it can become especially problematic when it is systematic
To prevent attrition from effecting the data scores of those who left should be removed
What is a Testing Threat and what can be done to avoid it?
A threat to internal validity in which participants results on a test are effected by having completed the test multiple times
To prevent testing threats a study could either use a posttest only design, different tests for the pretest and posttest, a comparison group, or counterbalancing
What is an Instrumentation Threat and what can be done to avoid it?
A threat to internal validity which occurs when an instruments accuracy changes over time
This can be avoided by confirming equivalence of tests or using a posttest only design
What type of replication involves repeating an original study with the same method, materials, variables, and hypothesis but a different sample
Direct Replication
What type of replication involves repeating an original study with the same method, materials, and hypothesis but a different sample and operationalize the variables differently?
Conceptual Replication
What type of replication involves repeating an original study with the same method, materials, variables and hypothesis but also adda new variable
Replication with Extension
What is the file drawer problem?
The idea that a meta-analysis might be overestimating the true size of an effect because null effects have not been included in the collection process
What does the p-value tell us?
The probability of getting results by chance
I.e. a p-value of .05 means there is only a 5% chance that the association found in the sample comes from a population in which there is no association
What does a t-test tell us?
Whether the difference between two groups means in an Independent-group design is statistically significant (Used when SD is unknown and sample size is small)
What does an f-test tell us?
Determines the equality of the variances of two data sets - the ratio of systematic variability to unsystematic variability (comparing two populations variances, used when Variance SD^2 is unknown)
What is a Factorial Design and what is the simplest version?
An experimental design involving two or mote independent variables
Simple Factorial Design: 2 x 2 (two independent variables each with two levels)
What is a main effect?
The possible effect of each independent variable with no relation to the other variables in the study
What is an interaction?
The possible changes that can occur due to interaction of variables (one variable changes as a result of another variable in the study)
What is more important, a main effect or an interaction?
Interaction
What is a Quasi-Experimental Design?
A design similar to an experimental design without the factor of random assignment (i.e. researcher loses full experimental control)
What is a Small-N Design?
A study which focuses on a single case (or a select few) usually done by measuring the dependent variable multiple times at each level
What is ecological Validity?
An aspect of external validity which is focused on whether a laboratory study generalizes to real-world settings (would the same effects happen in real-world situations or are they limited to a controlled lab?)
What is it called when laboratory research is just as realistic as in the real world?
Experimental Realism
What type of study occurs in the real world?
Field Studies
What does WEIRD stand for?
Western Educated Industrialized Rich Democratic
In testing the effect of noise on performance of a task, Ernie tests half of the participants in a cool, comfortable, quiet room, and the other half in a hot, noisy room. Ernie’s failure to control for temperature of the room may result in a __________
Confound
A study in which variables can’t be separated into levels or manipulated, they can only be measured to compare how they relate to one another
Correlational study
When a confounding variable is present in an experiment, one cannot tell whether the results were due to the __________ variable or the __________ variable
Independent variable
Confounding variable
If you measured students level of stress and the proximity (in days) to their first exam, what statistical test would you conduct?
Correlation
A researcher conducts his study with a small number of participants in order to reveal whether the instructions and procedures are clearly understood – this procedure is an example of a _______ study
Pilot
A researcher is trying to see if people’s favorite color (red, blue, or green) impacts their attitudes towards sports teams (Canadians, Leafs) – How many possible main effects are there? How many possible interactions are there?
Two Main Effects
One interaction
When a measure is not related to other variables that it theoretically should not be related to – this measure is said to have ______ validity
Discriminant
What is split half reliability?
Administering a test and then splitting in half and comparing results for both half
What is Alternate Form Reliability?
Administering a test in two halves at separate times
