Reading...
Play button
Play button
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;
30 Cards in this Set
- Front
- Back
|
Theory VS Hypothesis
|
a. Theory: A model we use to explain what we observe in the world around us (e.g. Theory of Gravitation, Evolution
An abstract thought or speculation; Arises from repeated observation and testing and incorporates Hypothesis: : an educated guess; A specific, testable prediction about what you expect to happen in your study. A clear, concise, testable statement A theory predicts events in general terms while a hypothesis makes a specific prediction about a set of circumstances. A hypothesis is a guess that has yet to be tested while a theory is/has been tested and is generally accepted |
|
|
Descriptive, comparative, correlational, quasi-experimental, and true experimental research
|
a. Descriptive research: Describing the way things are (sports, statistics)
b. Comparative research: Comparing two groups against one variable c. Correlational research: Trying to find a relationship between two variables (e.g. is there a relationship between height and weight of males at Vanderbilt University?) d. Quasi-experimental research: Non-random assignment of subjects to conditions (e.g. all girls go to left, all guys go to right) e. True experimental research: Random assignment of subjects to conditions |
|
|
Descriptive and inferential statistics
|
a. Descriptive: Describes the way things are (i.e. mean, median, mode, standard deviation, etc.)
b. Inferential: Trying to estimate magnitude of error between value of sample and value of population (e.g. statistical inference, T-test, P value) |
|
|
Conceptual and operational definitions of variables
|
a. Conceptual: One you find in a dictionary; the word-for-word definition of a word (e.g. Motivation is an individual’s willingness to engage in something)
b. Operational: Defining a variable in terms that you can measure (e.g. Teachers may measure a student’s motivation by how many minutes s/he spends in class |
|
|
Independent, dependent, and extraneous variables
|
Independent variable: one you can manipulate
Dependent variable: one that you measure based upon the value of the independent variable Extraneous variable: pesky things that can go wrong with the study, temperature, humidity, surface roughness. Other things that have an affect on dependent variable in ways you do not like. |
|
|
Measurement scales: nominal, ordinal, interval, and ratio
|
a. Nominal: Named categories (e.g. race, gender, religion)
i. Cannot measure mean, must use BAR GRAPH and MODE b. Ordinal: Categorical data that has an order (e.g. Kindergarten, 1st grade, 2nd grade, 3rd…) i. Cannot measure mean, must use BAR GRAPH and MODE c. Interval: Equal intervals on (e.g. temperature) i. Can use mean, median, or mode → must use LINE GRAPH or HISTOGRAM d. Ratio Scales: Scales that have equal intervals absolute zero (e.g. I am twice as tall as Jonny; height, weight, time, scores, income, etc.) i. Can use mean, median, or mode → must use LINE GRAPH or HISTOGRAM |
|
|
7. Measures of central tendency, variability, and how to treat outliers
|
a. Measures of central tendency: Single best score to calculate in a group of data (e.g. mean, median, mode)
b. Variability: How much spread there is in the data (i.e. variance, range, standard deviation, etc.) c. Standard Deviation: The average distance from the mean d. Outliers: Data points that are unusual, make us think whether to keep them or take them out; MUST EXPLAIN OUTLIERS BEFORE DISCARDING THEM → CIRCLE AND LABEL “OUTLIER” AND EXPLAIN IT |
|
|
8. Correlation coefficient (including estimating the size and value…and the effect of outliers)?
|
a. Correlations: Relationships; can be used to PREDICT things (e.g. your weight may be used to predict your height if there is a strong correlation between the two)
i. There is a correlation between any 2 variables, whether small or large. In order to get correlation between any two variables, mark them on a scatter plot and run a regression line through them 1. Basketball shaped line: correlation = 0 2. Football shaped line: correlation = .35 3. Corn on the cob shaped line: correlation = .75 |
|
|
9. Institutional Review Board (human participant protection): autonomy, beneficence, justice
|
a. Institutional Review Board (IRB): A group of institutions that receives federal funding to review experiments to determine if they are ethical (human participant protection); practices three terms:
i. Autonomy: Respect for the person 1. Strategy to ensure this → Ensured consent form ii. Beneficence: Ensuring a return on investment; cost-benefit tradeoffs iii. Justice: Ensuring that those participating in the study realize the benefits; must equally distribute the burdens and benefits of research |
|
|
10. Words of estimative probability?
|
Precision in language; how to convey the likelihood that something will happen (Don’t use WEASEL words that give no indication of likelihood – e.g. might, may, could, etc.)
|
|
|
11. Reliability and validity
|
a. Reliability: Consistency of measurement; a necessary but insufficient condition for validity
i. If the test is not reliable, it cannot be valid for any purpose b. Validity: How meaningful, useful and appropriate the inferences (conclusions )are (e.g. the eye chart test is a valid test FOR THE PURPOSE of determining whether or not one needs glasses) |
|
|
1. What are the ways of knowing we discussed in class? Why are some ways better than others? What’s the risk in believing something simply because it’s obvious?
|
a. Personal experience (seeing is believing, believing is seeing), authority (wrong all the time), logic/common sense, tradition (good, but things change over time), and systematic inquiry.
i. Logic: Of all the ways we can come to know the world, systematic inquiry is the best because it involves defining things very carefully, using carefully controlled studies that reduce errors and bias. People tend to believe anything they hear about human behavior, whether its true or not |
|
|
2. What’s the difference between correlation and causality? What’s the benefit of correlational data? What is required to present a strong case for a causal relationship between two variables?
|
a. Correlation: A way that allows us to understand the relationship/strength between two variables. Predicting one variable based on another. Yet, doesn’t tell us if A caused B or B caused A
b. Causality: MUST CONDUCT AN EXPERIMENT to establish a strong argument for causality (NOT a link or relationship between two variables) 2. Correlation doesn’t neccerily lead to causality (coffee -> cancer (caffeine and cigs, cigs actually cause) |
|
|
3. What is required before you can say you truly know something? Provide some examples.
|
a. Darwin Hunt (The Concept of Knowledge) says it has to be true, you have to be confident, you must have the right to know – this comes through evidence
b. Must KNOW it, must be CONFIDENT that you know it. c. JUSTIFICATION, comes from the amount of evidence you have d. Confidence should be tempered by how much evidence you have |
|
|
4. What are some key Do’s and Don’ts when graphing quantitative information
|
a. No color to differentiate categories in graphs
b. Don’t use 3-D c. Use bar graphs over pie charts to graph categorical data d. Make sure graphs are consistent in terms of font, color, etc. e. Don’t use lots of gray and lines in tables, this clutters them f. Use horizontal text for categories and chart labels |
|
|
5. List the required elements of qualitative and quantitative research questions and be able to provide examples of both.
|
a. Qualitative
i. What is the phenomenon? ii. Who are the participants? iii. What is the setting? b. Quantitative (e.g. what is the effect of consuming Gatorade at halftime on Vandy football players’ stamina at the end of the game?) i. What are the variables (consumption of Gatorade, stamina at the end of the game)? ii. What is the relationship between them? (Causal) iii. Who are the subjects? (e.g. football players at Vanderbilt) |
|
|
Differences between Quantitative and Qualitative research
|
Quantitative:
Specific Closed Static Outcome oriented Variables Qualitative: General Open Evolving Process oriented No variables I.E What are the attitudes of 8th graders at a school regarding the use of drugs who were taught drug ed |
|
|
3 types of research questions?
|
Descriptive
Correlational/comparative Experimental/Causal |
|
|
What are internal and external validity? Provide definitions and examples
|
a. Internal Validity: The extent to which the independent variable, and not other extraneous variables, produced the observed effect on the dependent variable )The extent to which the change of the independent variable change the dependent variable)
b. External Validity: The extent to which the results of the study can be generalize to other subjects, settings, and time i. E.g. can’t generalize a study from the 1950s about drug use to now |
|
|
6. What are major similarities and differences (5 ea.) between qualitative and quantitative research?
|
a. Similarities:
i. Both forms of systematic inquiry (ways of understanding the world) ii. Both have commonly accepted ways to conduct them iii. Both have common forms of reporting iv. Both are subject to IRB approval v. Both produce results that are tentative (e.g. sean replicates jim’s study and it turns out that sean’s is better) b. Differences (see handout) i. Commonly used terminology 1. Quantitative: hard data, statistical 2. Qualitative: field research, ethnography ii. Key concepts 1. Quantitative: variables, validity, hypothesis, testing, statistical significance 2. Qualitative: meaning, understanding, context iii. Research design 1. Quantitative: highly structured, formal, specific (e.g. pretest, posttest, experimental, control group design) 2. Qualitative: flexible/unstructured designs which evolve over course of research (e.g. plan on studying about effects of marijuana/ecstasy on students but find that Adderall is more pertinent to study, so switch to studying that) iv. Subjects/participants being sampled 1. Quantitative: involve many subjects representative of groups from which they are chosen. Ideal approach is random sampling/assignment (numbers) 2. Qualitative: Involve few subjects all w/ specific characteristics of interest to researchers. Subjects aren’t necessarily representative of groups of which they are chosen. Ideal is purposeful sampling (e.g. interview, observe/report → become part of environment you are researching) v. Researcher’s roles 1. Quantitative: detached, objective observers of events 2. Participant observers; observe and report from subject’s perspective (must develop close relationship with participants) |
|
|
7. How are sampling error, sampling fluctuation, and statistical significance related? How can you reduce sampling error and increase the likelihood your results will be statistically significant?
|
a. Sampling error is inevitable (when I measure a variable for a sample, I know that the average value of the sample will be different than the average value of the population); eliminate sampling error with a census, minimize sampling error by increasing sample size
b. Sampling fluctuation is also inevitable; both sampling fluctuation and sampling error are significant indicators of statistical significance c. **Statistical significance: Groups are different =, things are larger than you would expect from sampling fluctuation alone |
|
|
8. Calculate and interpret z-scores and effect sizes given basic data. Explain these concepts to your Uncle Bob .
|
a. Z = (individual score – group score) / standard deviation of the group
b. Effect size: (the average of the experimental group – average of control group) / standard deviation of control group 8. Z score, raw score – mean / standard deviation, Effect Size = mean of experimental group – mean of control/ SD (tells how powerful data is |
|
|
9. Be able to write an APA 6th edition citation for a peer-reviewed journal publication
|
Author, A. A., & Author, B. B. (2006). Title of article.
Title of Journal, volume number, nn-nn. • Example: Wright, S. P., Horn, S. P., & Sanders, W. L. (1997). Teachers and classrooms context effects on student achievement: Implications for teacher evaluation. Journal of Personnel Evaluation in Education, 11, 57-67. |
|
|
10. ****15 POINT QUESTION: What are statistical significance, effect size, and practical significance? (Be able to explain these concepts in terms your Uncle Bob, who doesn’t know statistics, would understand.)
|
a. Statistical significance: A test of statistical significance is a mathematical test that gives a yes/no answer to the question “Are the differences we observe larger than we would expect than through sampling fluctuation alone?” (i.e. “Hey, theres a statistical significance of the Vanderbilt football game)
b. Effect Size: Tells us the magnitude and the direction of that difference (e.g. Vandy football beat Eastern Michigan by 28 points) c. Practical Significance: Tells us how meaningful the significance is to humans (e.g. Vandy played great on parents’ weekend, so now the parents will come to more games) i. Takes into account ost, ethics, practicality Practical significance: important for mom and dad to see me win, went to national championship, now I’m going to the Superbowl. Tells us something effect size and stat sig how meaningful it is to people who have values take into account, cost ethics, practicality. |
|
|
12. Provide a detailed explanation of a pretest-posttest control group experimental design
|
a. Draw the diagram given in the handout
b. We do a pretest to do c. Control group reduces the threats of internal validity (?) d. Delayed retention test tells us how long the treatment persists e. Be able to explain what each component of the experimental design does |
|
|
13. List the 11 threats to internal validity discussed in class. Name one, define it, give a real-world example of how it could affect an experiment, and provide 3 strategies to reduce it as a threat.
|
a. Statistical regression of mean
b. Confusion c. Demoralization…………All slides are posted on website d. HERMIT’S DRED*** History Experimental mortality statistical Regression to the mean Maturation Instrumentation Testing Selection Diffusion compensatory Rivalry compensatory Equalization Demoralization |
|
|
14. Name (and describe with a real-world example) five ways you can you increase the reliability of your observations.
|
a. Target specific behaviors (e.g. how sharp the dancers’ turns are)
b. Use low inference measures c. Use multiple observers (e.g. use 5 trained judges instead of 1) d. Have observers be blind to conditions → unknowing to groups, hypotheses, etc. |
|
|
Goals of sampling
|
Avoid error and bias
Sample error: difference between true result and observed result, attributed to samples Bias: differences attributed to researcher |
|
|
Quantitative sampling strategies
|
Simple Random (random number)
Systematic Sample (every nth number) Stratified Sampling (proportions Cluster (naturally occuring groups are slected first) Convience (educationL |
|
|
Differences between Quantitative and Qualitative research methods
|
Common Themes:
Quantitative= Hard stats Qualitative = Field Research Goals Quan: Hypothesis Testing Qual: Meaning Design Quan: Structured Qual: Flexible Role Research: Quan = Detached Qual= Involved Participants Quan= Representative Qual= Selected |
