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;
130 Cards in this Set
- Front
- Back
|
What is science based on? |
Empirical evidence |
|
|
What is belief based/authority based on? |
authority |
|
|
Is science or belief based/authority tentative-theory based? |
Science; belief based is absolute |
|
|
What happens to science when evidence counters? |
Theory is questioned |
|
|
What happens to belief based research when evidence is countered? |
Evidence is questions
|
|
|
What is a hypothesis? What are defying characteristics it must have? |
An assertion about what is true in a particular situation; often, a statement asserting that two or more variables are related to one another. • must be testable • Must be falsifiable • Always tentative |
|
|
What is a theory? |
A systematic, coherent, and logical set of ideas about a particular topic or phenomenon that serves to organize and explain data and generate new knowledge. |
|
|
What is correlation? (positive, negative, none) |
A relationship between variables. As one variable changes, the other variable changes in a predictable way: • + correlation = Variables move in the same direction • - correlation = Variables move in opposite direction • 0 No relationship, No correlation. Closer to 0, either + or – is weaker |
|
|
What is causation? |
The change in one variable is causing a change in another variable. |
|
|
How are correlation and causation related? |
If there is a causal relationship between variables, there is also a correlation (there is a relationship between the variables. As one changes, the other changes in a predictable way. However correlation does not equal causation. |
|
|
Name the two main variables in correlational research? |
1. Predictor Variable 2. Criterion Variable |
|
|
Define predictor variable: |
A variable that is used to make a prediction of an individual's score on another variable (the criterion variable). However, this does not mean it is causing changes in the other variable. Conceptually similar to an independent variable. |
|
|
Define criterion variable? |
The variable/score that is predicted based upon an individual's score on another variable (the predictor variable). Conceptually similar to a dependent variable. |
|
|
Do we use manipulation in correlational research? |
No |
|
|
Can we determine causation from a correlational study? |
No |
|
|
What type of conclusions can we make from a correlational study? |
1. Descriptive conclusions 2. Predictive conclusions |
|
|
What are descriptive conclusions? |
Descriptive conclusions: Description of what trends are present in the current data i.e. as participants exercise more, they have higher GPA’s (example of positive correlation) |
|
|
What are predictive conclusions? |
Predictive conclusions: Predicting how variables will behave in relationship to each based i.e. if we know how much two students exercise, we can reliably predict who should have the higher GPA |
|
|
How do correlational studies differ from experimental studies? |
• We don’t manipulate the variables of interest in correlational studies but we do in experimental • We don’t really split up the participants into groups in correlational studies as we do in experimental studies • Each participant is their own plot point in correlational studies so we use scatter plots |
|
|
What type of research does directionality problem and third variable problem apply to? |
Correlational studies |
|
|
What is Directionality Problem? |
We don’t know if variable A is causing variable B or if variable B is causing variable A. Just one reason why we can’t make causal conclusions in correlational studies. |
|
|
What is Third Variable Problem? |
We don’t know if there is another variable that is causing both variable A and B. |
|
|
Name the two main variables in experimental research. |
1. Independent Variable 2. Dependent Variable |
|
|
Define independent variable? |
• must manipulate it (meaning we must control the levels to create at least 2 distinct condition) • must have at least 2 levels; actively controlling the levels and assigning participants |
|
|
Define dependent variable? |
Outcome variable based on IV |
|
|
What are 3 ways to measure DV? |
1) self report 2) Behavioral 3) physiological |
|
|
Define Self report? |
This method is usually the easiest. You can also ask about variables that might be much more difficult to measure in any other method, if not impossible. However, also usually more unreliable, people have bad memories, lie for whatever reason, or sometimes are not very self-aware. |
|
|
Define behavioral? |
Anytime a rater is observing the behavior of participants. It might be harder to come up with a valid behavioral method than self-report, but it’s usually more reliable. |
|
|
Define physiological? |
A body measure such as heart rate, blood pressure, fMRI, etc. This is often the most objective method, |
|
|
WHAT IS NEEDED FOR A GOOD EXPERIMENT? |
1. Must manipulate an I.V. with a minimum of 2 levels. 2. Must assign participants to I.V. through random assignment. 3. Must measure D.V. in 1 of 3 ways: Self-Report, Physiological, Behavioral (thru observation) 4. Must control extraneous variables. |
|
|
Define BTWN subjects design and give pros and cons. |
Participants are only in one condition. Different groups in each condition. PRO: Controls for order, sequence, carry-over, fatigue, practice effect. Needs less controls and therefore easier to design. Easier to use deception. CON: Needs more participants. Results less powerful because more susceptible to error variance/ individual differences because different groups of participants in each condition/level. |
|
|
Define W/IN subjects design and give pros and cons. |
Subjects participate in all conditions. PRO: More powerful, everyone is their own control, minimizes error variance. CON: More susceptible to order, carryover, practice, and fatigue effects. Harder to design because must control more for these problems. Also, if design needs deception, it may be more difficult to repeatedly deceive participants through multiple conditions. |
|
|
What is another name for Between and Within subjs design? |
BTWN = Independent Groups Design W/IN = Repeated Measures Design |
|
|
What are extraneous variables? |
Anything not IV or DV |
|
|
What a confounding variable (how does it behave)? |
Type of extraneous variable |
|
|
Why is confounding variable bad for an experiment? |
It varies systematically: has a different level for each condition, changes when the IV changes, confounded with the IV, behaves like the IV |
|
|
How is confounding variable related to extraneous variables? |
Alternative explanations in a study that we cannot eliminate because it is confounded with the IV (meaning the IV and the confound change at the same time) |
|
|
Define experimental control: |
Controlling extraneous variables to prevent them from becoming alternative explanations for the hopefully causal relationship between the IV and the DV. |
|
|
What are 3 WAYS TO CONTROL EXTRANEOUS VARIABLES? |
constant, randomization, counterbalance |
|
|
What does it mean to hold a variable constant? |
Keeping an extraneous variable the same for all conditions |
|
|
What does it mean to randomize a variable? |
Controlling for the effects of extraneous variables by ensuring that the variables operate in a manner determined entirely by chance. |
|
|
What does it mean to counterbalance a variable? |
A method of controlling for order effects in a repeated measures design by either including all orders of treatment presentation or randomly determining the order for each subject. |
|
|
How is randomization most often utilized in experimental research? |
Random Assignment. Determines how participants are assigned within the study. |
|
|
How is randomization used for a between subjects vs within subjects experiment? |
• For between subjects design, we randomly assign participants to conditions. • For within subjects design, we randomly assign participants to particular order of conditions. |
|
|
Are reliability and validity the same concept? |
• If something is reliable, it is not necessarily valid. If something is consistent, it is not necessarily true, we could consistently get the wrong answer. • If something is valid, it must be reliable. If something is true/accurate, it should be consistently accurate. |
|
|
What helps and hurts each type of validity? |
Validity is truth/accuracy |
|
|
What is internal validity? |
The manipulation of the I.V. is causing the D.V. We must manipulate I.V., control for extraneous variables, and measure the D.V. Important to control for extraneous variables. |
|
|
What is terrible for internal validity? |
Confounds are terrible for internal validity and make it impossible for us to make causal conclusions. |
|
|
What is external validity? |
If findings of an experiment are generalizable, meaning they can be applied to the real world. |
|
|
What must you have for external validity? |
1. a representative sample and 2. conditions that could reasonably resemble the real world. |
|
|
Our findings must have _____ validity in order to generalize those findings. |
internal |
|
|
What is construct validity? |
Accurately measure the construct you want to measure. |
|
|
What ensures good construct validity? |
Good operational definitions ensure good construct validity. |
|
|
What is reliability? |
consistant |
|
|
What is Test-Retest Reliability? |
Getting the same results over and over having taken/administered a test/measure. A reliability coefficient determined by the correlation between scores on a measure given at one time with scores on the same measure given at a later time. Ex 1. Taking the SATS over and over and getting the similar scores Ex 2. Using the same thermometer and getting the same temperature. |
|
|
What is Inter-Rater Reliability? |
Different raters/observers coming up with the same ratings/observations within an experiment. |
|
|
What helps inter rater reliability? |
train observers with good and specific operational definitions |
|
|
What is Replicability? |
Being able to repeat an experiment with the same results, usually by a different investigator. |
|
|
What are problems with research? |
1. EXPERIMENTER BIAS 2. SUBJECT BIAS 3. DEMAND CHARACTERISTICS 4. Material bias |
|
|
What is experimenter bias? |
Any intentional or unintentional influence that the experimenter exerts on subjects or data to confirm the hypothesis under investigation. |
|
|
What is subject bias? |
When participants affect the study in a negative way by forming expectations about the hypothesis of the study, and act to confirm or disconfirm the hypothesis or using self selection effect |
|
|
What is demand characteristics? |
Cues that inform the subject how he or she is expected to behave. • If the demand characteristics occur in the materials, it is a type of material bias. • The experimenter can also be the source of the demand characteristics |
|
|
What is material bias? |
A problem with the materials of the study |
|
|
What are types of material bias? |
1. floor effects 2. ceiling effects 3. specific item effects |
|
|
What is flood effect? |
All scores are at the bottom. Usually due to failure of a measure to detect a difference because it was too difficult. A type of material bias |
|
|
What is ceiling effect? |
All scores are at the top. Usually due to failure of a measure to detect a difference because it was too easy. A type of material bias |
|
|
What is specific item effect? |
A type of material bias where the materials systematically vary • When the confound is in the materials |
|
|
Define the term variable: |
Any event, situation, behavior, or individual characteristic that varies—that is, has the potential to take on at least two values |
|
|
What are the types of variables? |
1. Qualitative 2. Quantitative 3. Discrete 4. Continuous 5. Dichotomous 6. Concrete 7. Abstract |
|
|
Define dichotomous variables? |
Only 2 options: male or female |
|
|
Define concrete variables? |
Directly and readily measurable. |
|
|
Define abstract variables? |
Not directly measurable, i.e. emotions. Must use operational definitions to turn abstract variables into concrete variables. |
|
|
What are operational definitions? |
The specific procedure or method that is used to observe, measure, and/or manipulate a variable. Operational definitions make abstract variables into concrete ones. |
|
|
What is ordinal scale? |
Rank order: 1st, 2nd, 3rd, etc. Examples: outcomes of a race, ranking of sports teams |
|
|
What is interval scale? |
Equal intervals between consecutive points of the scale. Arbitrary zero. Examples: Fahrenheit scale, Likert scales |
|
|
What is ratio scale? |
Equal intervals between consecutive points of the scale. Meaningful zero. Examples: # of children, where zero children actually means no children; Kelvin Scale; time |
|
|
What is descriptive stats? |
Statistical measures that describe the results of a study; descriptive statistics include measures of central tendency (e.g., mean), variability (e.g., standard deviation), and correlation (e.g., Pearson r). |
|
|
What is Inferential Statistics? |
Statistics designed to determine whether results based on sample data are generalizable to a population (F statistics, p values) |
|
|
What is Statistical significance: |
Rejection of the null hypothesis when an outcome has a low probability of occurrence (usually .05 or less) if, in fact, the null hypothesis is correct. |
|
|
When would you use a t-test? |
When a design has only 1 IV with just 2 levels. |
|
|
When would you used an F-test? |
When a design has more than 2 conditions. So 1 IV with 3 or more levels or multiple IV’s. |
|
|
Define systematic variance and give the other name for this concept |
(between group variance)Variability in a set of scores that is the result of the independent variable (when there is no confound); statistically, the variability of each group mean from the grand mean of all subjects. |
|
|
Define error variance and give the other name for this concept. |
(within group variance) Random variability in a set of scores that is not the result of the independent variable. Statistically, the variability of each score from its group mean. |
|
|
WHAT IS OBSERVATIONAL RESEARCH? PRO AND CON? |
Categorize degree of observation in regard to the level of interaction between observer and observed. Try to observe behavior as it naturally occurs, trying not to affect.PRO: Findings can be more generalizable.CON: Lacking control. |
|
|
Define Naturalistic Observation/ Field Work/ Non Reactive/ Unobtrusive? |
The observer is not interacting with subjects. Trying to keep yourself, as observer, as unobserved by subject as possible. Two different types from lecture: A. Unseen (unobtrusive): don’t know they’re being watchedExample: physical trace measures: such as looking at the fingerprints at a museum to see which exhibits are being visited the most B. Seen: more potential for subject reactivity but people being watched will get used to your presence and act more normally again (habituation) |
|
|
Define Participant Observer? |
Occurs when observer must join the observed group to study, but attempts to not affect anything. Not a good idea if observer is already member of the group, lacks necessary objectivity. Pros: Better understanding of situation; deeper insights; might have access to info and situations that otherwise couldn’t get access to |
|
|
Define Quantitative Research? |
Tends to focus on specific behaviors, usually from larger samples, that can easily be quantified/ counted and then analyzed using statistical analysis. One can even use non-numerical answers to questions that can easily be assigned numerical values and then statistically analyzed. |
|
|
Define Qualitative Research? |
Focuses on people behaving in natural settings and describing their world in their own words, usually focusing on a small group within a limited setting. Conclusions of qualitative research are based on interpretations drawn by the investigator. Qualitative because expressed in non-numerical terms using language and images. |
|
|
What are concerns for observational study? |
1. OBSERVATIONAL BIAS 2. SUBJECT REACTIVITY |
|
|
What is observational bias? |
An observers cognitive biases can affect what they see; unconsciously influence observations: how they observe, what they see, how they look at the data. Especially when they know the hypothesis, they’ll look for what they expect to see. |
|
|
How do you control for observational bias? |
1. Have multiple observers 2. Blind observer 3. Narrow focus and use specific directions for observations: systematic observation to make observing as objective as possible |
|
|
What is SUBJECT REACTIVITY? |
Subjects change their behavior because they know they are being observed. Observers, therefore, must try to be unobtrusive. |
|
|
What is individual sampling? |
Only observe predetermined individuals/ groups and not others. |
|
|
What is time sampling? |
Set time intervals to observe and not observe. This is predetermined. Prevents observer seeing what they want to see. Only record event occurrence during predetermined time intervals. |
|
|
What is FREQUENCY METHOD? |
How often does behavior/event take place/occur |
|
|
What is interval method? |
Divide time into intervals: say every 30 seconds. Did the event/behavior occur during each time interval? Could be yes or no. To quantify: for how many of the intervals did the event take place? Could also use those time intervals along with frequency or duration. Example: For 30 minutes of observation, we looks at intervals of 5 minutes. For each of the 6 |
|
|
What is duration method? |
How long does observed behavior last? The duration of the observed behavior/event |
|
|
What are OTHER TYPES OF NON-EXPERIMENTAL DESIGNS? |
case, archival, survey |
|
|
What is case study? Pro and Con? |
Looking at particular situations. Often only individual or small group. Could also be a specific neighborhood or business. PRO: in depth information; can study very rare cases and/or situations that can’t be replicated in experiment; could be a starting off point for researching a phenomenon CON: Low external validity due to small sample size so hard to generalize; No internal validity |
|
|
What is archival study? Pro and Con? |
• Using existing data that’s already been collected/existing records; doesn’t have to be old • PRO: Less time consuming; cost effective; convenient way to study changes over time (without have to use a longitudinal design) and over space/geographical regions • CON: Issues of access (gov’t, medical records) data might not exist; can’t always be certain that original investigators used good methods; current researchers didn’t actually collect data; (issues of instrument decay) |
|
|
What is survey? Pro and con? |
Asking questionnaires, depends on self-report. • PRO: Cheaper, easier, can obtain a lot of info at once, can ask about things that would be difficult or impossible to observe • CON: Hard to ask good questions that are not biased; self-report is relatively unreliable (ppl lie, forget, don’t take it seriously, social desirability); Non-response bias |
|
|
What is Non-response bias? |
the people that do respond to surveys are not necessarily representative of the entire population because they may be inherently different than those who do not respond to surveys |
|
|
QUASI-EXPERIMENTAL RESEARCH How does this differ from true experimental research? |
• Lacks random assignment (true manipulation) • CANNOT making causal conclusions |
|
|
What is Quasi-Independent Variable? ex? |
Not randomly assigning participants to levels of the variable • Examples: Gender, ethnicity, SES, age, smoking, illicit drug use |
|
|
What are Reasons why we’re not randomly assigning participants to groups in quasi independent variable? |
• Not possible • Unethical • Too difficult |
|
|
What is Quasi-Factorial Design? |
• At least 1 quasi-IV (don’t need a true IV for this design, could be all quasi-IV’s, could also be 3 true IV’s and only 1 quasi-IV and still quasi-factorial design) • at least 2 IV’s (either quasi or true) |
|
|
When can we say yes or no to causal conclusions for quasi-factorial design? |
• main effect of quasi-IV: no causal conclusions • main effect of true-IV: yes causal conclusions • interaction between quasi-IV and true-IV: no causal conclusions |
|
|
What is Pretest-Posttest? |
time one - event - time two - event data taken at different times |
|
|
What are threats to pretest posttest? how to solve? |
history, testing effect, instrument decay, maturation (if study is over extended period), mortality (the longer the study, the bigger the issue of mortality), selection effects, (regression to mean, not the biggest issue here but might possibly be a problem) solve: non equivalent control group |
|
|
What is non equivalent control group? |
Called Non-equivalent due to lack random assignment, which means I’m not controlling for participant variables = individual difference between the participants T1 ------ Event/Manipulation-------------T2 T1 ------Control (No event/man.) -------T2 Helps to control for: maturation, testing, some history effects (only makes selection effects worse |
|
|
What is interrupted time series? |
T1-T2-T3--Event/Manipulation--T4-T5-T6 Almost the same as pretest-posttest, except here there’s multiple pretests and multiple posttestsBiggest Threats: history, selection effects, (potentially maturation, mortality, instrument decay) |
|
|
Threats to interrupted time series? |
Biggest Threats: history, selection effects, (potentially maturation, mortality, instrument decay) |
|
|
What is Prospective Ex Post Facto? |
Starting with the “predictor”/suspected cause and follow groups along People who eat low amounts of sugar➔ less diabetes People who eat high amounts of sugar➔ more diabetes |
|
|
Biggest threat to ex post facto design? |
mortality, selection effects, |
|
|
What is Retrospective Ex Post Facto? |
Start with the outcome, split into groups based on the outcome, and look backwards at what I suspect is the predictor/ “cause” Start and look back Usually does rely on self-report, which the most unreliable: people lie, people misremember especially way in the past |
|
|
What is developmental research? |
Changes due to age or time will always be a variable of interest, we’re looking for maturation effects (therefore no longer a threat/alternative explanation, it’s what we hope to see) |
|
|
Wha tis cross sectional method? Threat? |
Comparing Different age groups (cohorts) at a single point in time, Between-subjects Threat: cohort effects |
|
|
What are developmental research? |
cross sectional longitudinal sequential |
|
|
Wha tis longitudinal method? threat? |
Within subjects, over time, comparing one cohort/one group the same age following over time ex. 10 year olds - 15 year olds - 20 year olds (say start 2015) (2020) (2025) Threat: secular trends/time of testing effects, mortality |
|
|
What is sequential method? (Cross sequential) |
A combination of cross-sectional and longitudinal techniquesFollow multiple cohorts/age groups through time. 2015 2020 2025 Start with 5 year olds - 10 year olds - 15 year olds Start with 10 year olds - 15 year olds - 20 year olds |
|
|
What is history effect? |
Any outside even that causes changes in people (at global/large scale or individual levels) |
|
|
What is maturation effect? |
Changes that happen as people age, happen over time, changes due to normal development |
|
|
what is testing effect? |
The test itself affects the outcome, affects results on future tests |
|
|
What is regression to the mean? |
Extreme scores, upon further testing, tend to become more average; statistical phenomenon |
|
|
What is selection effect? |
Happens when we don’t have random assignment(self-selection is a sub-type of selection effects, specifically when participants choose their groups) |
|
|
What is Instrument Decay? |
The method of measurement is changing, which could cause changes in the results (method of measurement could be survey, an actual instrument like a scale, test, or observer, etc.) |
|
|
What is cohort effect? |
Generational differences between age groups other than just normal developmentWe’re saying that because of cohort effects 20 year olds today are different than 10 year olds will be in 10 years. Type of selection effect because we’re not randomly assigning people to cohortsExample: 20 year olds and 40 year olds might be different because of the type of technology they were exposed to when they were children |
|
|
What is Secular Trends/ Time Of Testing Effect? |
Changes in people that are due to the changes over time of the socio-political-cultural context/environment |
|
|
FACTORIAL DESIGN Define. What are important characteristics of this type of design? |
• There must be at least 2 or more IV’s • Each level of 1 IV must cross with each level of the other IV (this is how we get the matrix/grid we’ve been using in session) |
|
|
Why do factorial design? |
• We can see interactions between IV’s. • We have better control by eliminating an extraneous variable. The second IV would have been an extraneous variable in a simple design, but instead we incorporate it into the design as another IV |
