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;
224 Cards in this Set
- Front
- Back
What is epistemology? What are the two main frameworks? What are others? |
Epistemology: Theory of knowledge Positivism: Knowledge comes from scientific fact Empiricism: Experience and senses define knowledge Social constructionism, feminism, queer theory |
|
How often do fake news headlines fool American adults? |
75% of the time |
|
How often did respondents who cited FB as a major news source feel that fake news headlines were accurate? |
83% |
|
Good research: |
1. Based on work of others 2. Can be replicated 3. Is generalise to other settings 4. Is based on logic and tied to theory 5. Is doable 6. Generates new questions 7. Is incremental 8. Is apolitical |
|
What is the hypothetico-deductive method? |
Formulating a hypothesis that can be proved or disproved through experimentation or observation |
|
Scientific method steps? |
1. Asking the question 2. Identifying important factors 3. Formulating a hypothesis 4. Collecting relevant information 5. Testing the hypothesis 6. Working with the hypothesis 7. Reconsider the theory 8. Ask new questions |
|
Basic vs applied research? |
Basic: Answers questions about the nature of behavior. Doesn't seek to solve a problem, no applications in mind. Strict research protocols. Applied: Less strict process. Takes place in real world. Addresses issues in which there are practical solutions. Difference: Both important, neither superior. Science depends on synergy between both. Applied research usually guided by basic research. |
|
N vs n? |
N: Total participants n: Participants in each group |
|
Types of non-experimental research? |
Descriptive, historical, correlational, qualitative |
|
Types of experimental research? |
Pre-experimental, true experimental, quasi-experimental |
|
What is descriptive research? |
Describes phenomena existing in the present. Provides a broad picture and serves as a basis for other types of research. |
|
What is historical research? |
Describes events that took place in the past. Uses primary and secondary sources |
|
What is correlational research? |
Asks what variables have in common. Does not ask whether one causes the other. |
|
What is qualitative research? |
Examines behaviour in natural social, cultural and political contexts. Usually does not use quantitative data |
|
Variable vs value? |
A value is a subset of a variable Variable: Height, weight, hair colour Value: 156cm, 91kg, black |
|
What is an independent variable? |
The variable that forms the groups/treatments/conditions in a study. Manipulated by the researcher directly or indirectly |
|
What is a dependent variable? |
The outcome, the thing being assessed or measured. Scores depend on the independent variable. Need to be operationalised |
|
What is a control variable? |
The variable other than IV that can affect the DV. Needs to be controlled or monitored |
|
What is an extraneous variable? |
Unaccounted for variable which can confound data by manipulating the DV and making it impossible to draw causal relationships |
|
What is a mediator variable? |
Explains the relationship between two variables E.g. Number of bystanders > DIFFUSION OF RESPONSIBILITY > Helping behaviour |
|
What is a moderator variable? |
Changes the strength between two variables E.g. Work stress < COPING STRATEGY > Drinking problems |
|
What is a between subjects design? |
AKA independent samples Each subject is exposed to one level of each IV |
|
What is a within subjects design? |
AKA repeated measures Each subject is exposed to all levels of each IV |
|
What is a hypothesis? |
"If... then..." statement that predicts the outcome of a study |
|
What is falsification? |
The process taken to demonstrate something to be false. Hypotheses can be falsified or rejected |
|
What is the null hypothesis? |
A statement of no difference, relationship or effect. Research assumes it to be true so it can be disproved |
|
Population vs sample? |
Population: The collection of units to which we want to generalise our findings Sample: Smaller collection from population that is used to infer characteristics about the population |
|
How do you get a more accurate representation of a population? |
Increase the sample size |
|
What is estimation or inference? |
Generalising results from a sample to a population |
|
What is statistical significance? |
The degree of risk that you are willing to take that you will reject a null hypothesis (error rate) Calculated as a p value |
|
What is a p value? When can we reject the null hypothesis? |
The probability that the data would be at least as extreme as those observed, if the null hypothesis were true If p < 0.05 we can reject the null hypothesis. |
|
What does Fisher and the Lady Tasting Tea tell us? |
Only when there is a very small probability that an observed behaviour is due to chance alone can we conclude that there is a genuine effect |
|
What is critical appraisal? What isn't critical appraisal? |
Is: A balanced assessment of process and results undertaken by all health practitioners Isn't: Dismissal of research, narrow critique of reults |
|
What is the purpose of literature review? |
Discovering what has been done and needs to be done Chronological representation of ideas Shows which ideas have been abandoned/confirmed |
|
What is a general source? |
An overview of a topic, leads to further information E.g. Newspaper, magazine |
|
What is a secondary source? |
'Once removed' from original research E.g. Textbook, encyclopedia |
|
What is a primary source? |
Original report of original work E.g. Journal article |
|
What is a journal? |
A collection of research articles published in a discipline. Most important primary source of information about a topic |
|
How many times more likely is an article to be published if results are significant? |
3 |
|
Which sections are found in a manuscript? |
Title page, abstract, introduction, method (participants, materials and procedure), results, discussion, references, appendices, author notes, footnotes |
|
Approximate word limit of a title and running head? |
Title: 12 words Abstract: 50 characters |
|
What is on a title page? |
Running head, title, author, affiliation |
|
What is an abstract? |
Approx 150-250 words, summary of report Purpose of paper, participants, design, findings, implications, conclusions |
|
What is an introduction? |
No heading, introduces problem, literature review, aim, hypothesis |
|
What is a method? |
How study was conducted, operationalise variables |
|
What is the participants section? |
Includes demographics, sub-groups, animal info, sampling, payments, ethics |
|
What is the materials and procedures section? |
Lists anything used to conduct study, within/between subjects, conditions, instructions given to participants |
|
What is a figure? |
Any diagram presenting data that is not a table. E.g. Charts, graphs Title goes at the bottom |
|
What is in the results section? |
Summary and analysis of data Include test statistics |
|
Where does the heading of a table go? |
Top |
|
What is in the discussion section? |
Whether or not hypothesis was supported Reflection on relationship to previous literature Limitations, implications, generalisability |
|
What is a reference? |
Acknowledges previous work Helps build a network of literature Hanging indent in APA 6 |
|
What is a DOI? |
Digital Object Identifier, permanent link to an article |
|
What is in an appendix? |
Supplementary information listed in alphabetical order E.g. raw data |
|
What is circular definition? |
Using a the word being defined in the definition E.g. A conversation is when you converse with someone |
|
What is plagiarism? |
Claiming someone's ideas as your own Not referencing correctly |
|
What is ethics? |
A code of what is morally right or wrong E.g. Integrity, competence, beneficence, responsibility, honesty, autonomy, privacy, confidentiality |
|
Utilitarianism vs deontology? |
Utilitarianism: Sacrificing something for the greater good. E.g. killing patients to find a cure for cancer that would save millions Deontology: We are bound to follow moral principles E.g. We can't kill those few people because their lives matter |
|
When was Milgram's obedience experiment conducted? |
Early 1960s after WWII |
|
When was the Tuskegee Syphilis Study conducted and by who? |
US Public Health Service 1932-1972 |
|
What prompted the Belmont Report? What is its purpose? |
The Tuskegee Syphilis Study To protect participants in clinical trials and research studies. Three principles: beneficence, autonomy (respect for persons) and justice. |
|
What is informed consent? |
Information that might influence whether someone participates in research. States the purpose of research, procedures, risks and benefits, compensation, confidentiality, permission to withdraw, contact information |
|
What is deception? |
Withholding information to a participant about the study |
|
What is debriefing? |
Provides an explanation for withheld information, occurs immediately after study if possible, makes sure participants don't leave with ill feelings toward the field of psychology |
|
Alternatives to deception? |
Role playing, simulation studies, honest experiments |
|
What are the commitments of a researcher? |
Implied contract with participants Punctuality Summary of details to participant Course credit Details that maintain trust between participants and researchers |
|
What does the Nuremberg Code state? |
Voluntary consent Good of society not unnecessary Results will justify performance No physical/mental suffering or injury No risk of death or disabling injury Never exceed degree of risk needed to solve problem Proper preparations Only scientists conduct Liberty to bring experiment to end Be prepared to terminate experience |
|
What does The Declaration of Helsinki state? |
Respect for individual Right to self determination Right to make informed decisions Investigator's duty solely to patient Subject's welfare must take precedence Ethical considerations must take precedence Increased vulnerability of individuals and groups calls for special vigilance |
|
What does The Institutional Ethics Review System State? |
Any institution receiving funds needs a Human Research Ethics Committee Responsible for reviewing research at the institution One member must be from outside All research by students/faculty/staff must be reviewed |
|
Three types of research in relation to ethics? |
Exempt research Minimal risk research Greater than minimal risk research |
|
What is a risk? |
A potential for harm, discomfort or inconvienience |
|
What does the assessment of risk involve? |
Identifying risks Gauging probably of minimising Determining whether they are justified Determining how they can be managed |
|
What are the main two risks of psychological research? |
Physical harm, stress, loss of confidentiality, |
|
What are the benefits of psychological research? |
Educational benefits Treatment Material benefits Personal satisfaction Benefits to society |
|
What does the APS Code of Ethics state? |
Respect for the rights of people and peoples including the right to autonomy and justice Propriety incorporating the principles of beneficence and non-maleficence and responsibility to clients, the profession and society Integrity reflecting the need for psychologists to have good character and acknowledge the high level of trust intrinsic to their professional relationships and impact of their conduct on the reputation of the profession |
|
What is fraud? |
Fabrication of data |
|
What is p-hacking? |
Changing experimental design and analysis to change to p-value |
|
What is measurement? |
The assignment of values to outcomes |
|
What are the four levels of measurement? |
Nominal: Differ in quality, not quantity (eye colour) Ordinal: Same but have an order (horse racing) Interval: Same but has equal intervals (temperature) Ratio: Same but scores are meaningful (weight) |
|
What is reliability and how do you increase it? |
True score / True score + error Increase number of items/observations Eliminate ambiguity Standardise conditions Minimise effects of external events Standardise instructions Standardise scoring |
|
How do you measure relibility? |
With Pearson's r or Pearson's product moment correlation coefficient. Shows how similar scores are from time 1 to time 2 |
|
What is test-retest? |
Measuring the same individuals at two points in time |
|
What is parallel forms? |
Different forms of the same test given to the same group of participants |
|
What is interrater? |
Proves reliability when multiple raters agree on same thing. Rater to rater not time to time |
|
What is internal consistency? |
Uses responses at only one time and fouses on consistency items |
|
What is validity? |
A test doing what it should |
|
What is face validity? |
A test being relevant to what it is intended to measure |
|
What is content validity? |
Content of the measure compares with the universe of content defining the construct |
|
What is criterion-related validity? |
The extent to which a score indicates a levelof performance or (externally set) criterion against which it is compared |
|
What is construct validity? |
The extent to which an assessment corresponds with other variables as predicted by theory |
|
Types of construct validity? |
Convergent validity: Correlate with questionnaires that measure the same or related constructs Discriminant validity: Should not correlate with questionnaires that measure different or unrelated constructs |
|
What is probability sampling? |
Likelihood of any one member of population being selected known |
|
What is nonprobability sampling? |
Likelihood of selecting any one member fromthe population is not known |
|
What is simple random sampling? |
Each member of the population has an equal and independent chance of being selected |
|
How do you use the table of random numbers? |
Use equal numbers of each digit, randomly select row, select two digit number and read down list With replacement: Duplicates Without replacement: No duplicates |
|
What is systematic sampling? |
Every kth name on a list is chosen, k = any number between 0 and the size of the sample |
|
What is stratified sampling? |
Groups divided into layers (strata) so that the sample/population is fairly represented |
|
What is cluster sampling? |
Selecting units of individuals rather than individuals themselves |
|
What is convenience sampling? |
Uses a captive audience E.g. uni students needing course credit |
|
What is quota sampling? |
Keep selecting individuals to fit each strata |
|
What is a measurement error? |
A discrepancy between the data found and the true value of the measurement |
|
Who did the Standord Prison Experiment and why? |
Phillip Zimbardo to test the power of the situation |
|
What is a test? |
A tool that measures behaviour |
|
What are some impotant test issues? |
How test will be administered Will it be standardised Nature of behaviour being tested |
|
What is an achievement test? |
Measures knowledge of a specific area, usually in learning Standardised: Produced by commercial publishers Research generated: Purpose built more specific Both of these can be norm-referenced (compare an individual's performance to others) or criterion-referenced (measures mastery not where a person sits relative to others) |
|
What does a multiple choice test consist of? |
A stem (posed question or problem) One correct option Distractors (wrong answers) |
|
Advantages/Disadvantages of multiple choice? |
Easy to score Assess any content Focus on knowledge not writing Good items may be reused Good distractors help in diagnosis Difficult to fake Restricts creativity Test anxiety Limits content assessed |
|
Tests - content vs process? |
Content: Attributes to be assessed Process: Ways to assess attributes |
|
What is item analysis? |
Assessing the value of each item to decide whether it should be retained or replaced |
|
Difficulty index vs discrimination index? |
Difficulty: amount of people who got it correct Discrimination: The amount of students in higher group who got it correct vs those in the lower group that got it correct As difficulty increases, discrimination is constrained |
|
What is an attitude test? |
Assesses an individual's attitudes, feelings, perceptions of something |
|
What is the Thurstone scale? |
First formal scale to measure an attitude Made up of statements with numerical values I believe the church is awesome I believe the church is okay I believe the church sucks |
|
What is a Likert scale? |
Approx 5-point scale with question and answers I like cats 5 Strongly Agree 4 Agree 3 Neutral 2 Disagree 1 Strongly Disagree |
|
What is a Guttman scale? |
Ordinal level Responses from weaker to stronger If you agree with the 2nd, you also agree with the first etc 1 I am willing to eat ice cream 2 I am willing to smell ice cream 3 I am willing to eat ice cream 4 I love ice cream |
|
Types of personality tests? |
Projective: Testee given ambiguous stimulus and asked to formulate a response (TAT) Structured: Use a set format of response (True/False) |
|
Which types of psychology use observation? |
Social, clinical, forensic, organisational, cognitive, developmental |
|
What is systematic observation? |
Using a coding system to categorise behaviour through unoccupied play, solitary play, parallel play and group play |
|
Advantages of systematic observation? |
Systematic Permanent record of data Extraneous variables controlled Replication Observer bias minimal |
|
Disadvantages of systematic observation? |
Lack of ecological validity Lack of behavioural spontaneity/realism |
|
What is naturalistic (field) observation? |
Observing people in their natural environment |
|
What is full participant observation? |
Research doesn't disclose their identity Pretends to be member of group |
|
What is participant as observer? |
Researcher's identity is not a secret, but kept quiet |
|
Advantages of naturalistic (field) observation? |
Ecological validity Useful when controlled setting isn't possible |
|
Disadvantages of naturalistic (field) observation? |
Difficult to conduct Difficult to be unobtrusive Poor control Observer bias Replication difficult Ethics |
|
Form vs consequence? |
Form could be jogging Consequence would be the result of jogging |
|
What is time sampling? |
Observing each member of a group in rotation for a certain amount of time |
|
How to calculate amount of behaviour? |
By amount spent on each behaviour regardless of when it occured |
|
What is the index of concordance? |
Agreements / Agreements + Disagreements |
|
What are jingles and jangles in a questionnaire? |
Jangle: Different labels for the same thing Jingles: Giving different constructs the same name |
|
What are the structures of interviews? |
Unstructured interviews Semi-structured Structured |
|
What are the types of probes? |
Cues, knowledge, analogues, goals, options, basis, experience, aiding, time pressure, situation assessment, hypotheticals |
|
Advantages/Disadvantages of interviews? |
Flexible Set tone and agenda Expensive No anonymity Lack standardised questions |
|
What are the 10 commandments of interviewing? |
Don't begin the interview cold, remember you are there to get information, be direct, dress appropriately, quiet place to conduct interview, don't give up, use a recorder, make the interviewee part off the interview, practice, thank interviewees and ask for questions |
|
How do you conduct survey research? |
Clarify objectives Identify sample Define method Coding Scoring |
|
Advantages/Disadvantages of survey research? |
Good generalisation Efficient data collection Can yield accurate results Interviewer/interviewee bias Non-reponse |
|
What is a scatterplot? |
Plots scores of one variable against scores of another |
|
What are the relationships that scatterplots show? |
Positive, negative and zero relationships |
|
What is tertium quid? |
When a third variable causes a relationship between variables |
|
What is the correlational coefficient and when is it considered weak/strong? |
+1 perfect positive -1 perfect negative 0 no relationship 5-8 said to be strong |
|
What are the coefficient of determination and alienation? |
Determination: Squared value of correlation coefficient Alienation: 1 - Coefficient of determination |
|
What is the symbol for Spearman's correlation coefficient of a population? |
p (rho)
|
|
What does p < 0.05 mean? |
There is less than a 5% chance that the sample comes from a population in which there is no relationship between the variables |
|
What is an outlier and how does it affect the data? |
Lies outside the other value in a data set Skews results and changes correlation |
|
What is regression? |
A straight line best fitting the scatterplot |
|
What is social constructionism? |
Understanding the ways in which people might develop and acquire knowledge through their social reality |
|
What are the two qualitative research methods? |
Those that emphasise meaning for participants (phenomenology, ethnography, grounded theory) Those that concentrate on language use (discourse analysis) |
|
How do qualitative and quantitative research relate to theory? |
Qualitative: Theory generation Quantitative: Theory verification |
|
Advantage/Disadvantage of a case study? |
Insight gained that can't be gained through experimentation Cannot generalise |
|
What is ethnography? |
Looks at group behaviour rather than individual behaviour Research immerses themselves in culture |
|
What is thematic analysis? |
Attempts to find patterns or themes in the data |
|
What is Foucauldian analysis? |
How statements/images construct the way a thing is thought about; words are objects. Focus on power relations |
|
What is grounded theory? |
Theory induced, generated and grounded in data |
|
What is interpretive phenomenological analysis? |
Explores the processes through which individuals make sense of their own experiences E.g. The grandchildren of Holocaust survivors saying the trauma is "always there" |
|
Which technology does qualitative research use? |
Programs aiding in analysing transcripts (Leximancer) or cataloguing of themes (NVivo) |
|
What is triangulation? |
Using multiple methods to understand a phenomenon Qualititative: Prevalence of depression in society Qualitative: What depression feels like |
|
Levels of data coding? |
1st: Textual data organised
2nd: Interpreted 3rd: Captures overall meaning |
|
What is a pre-exerimental design and what are some examples? |
No random assignment, no control group, little control over extraneous variables, less power to reveal causal relationships One Shot Case Study Design: Participants assigned to group > treatment given > post-test One Group Pre-Test Post-Test: Participants assigned > Pre-test > Treatment > Post-test |
|
What is a true experimental design and what are some examples? |
Uses random assignment, experimental/control groups. Pre-Test Post-Test: Random assignment > pre-test > no/treatment > post-test Post-Test Only: Random assignment > no/treatment > post-test Solomon Four Group: Random assignment to four groups > two pre-tests/two no pre-test > two treatments/two no treatments > post-test |
|
What is a randomised control trial? |
Tests treatment while removing bias. Control group receives standard treatment, placebo or nothing. Experimental group receives treatment being tested. |
|
What are some issues with evidence? |
Not everyone agrees on what good evidence is Socio-historical context Qualitative research |
|
What are some issues with RCTs? |
Only used in clinical psychology Low external validity Cannot capture psychotherapeutic experience Cannot study rare/unlikely events Ethics |
|
What do psychologists consider other than levels of evidence? |
Relevence of findings, applicability to other settings, strength of effect, magnitude of effect size, quality of methods used to remove bias |
|
What are the two types of validity? |
Internal: The accuracy in concluding that the outcome of an experiment is due to the IV External: The extent to which the results can be generalised |
|
What are threats to internal validity? |
History, maturation, selection, testing, instrumentation, regression, mortality, experimenter bias, demand characteristics |
|
What are threats to external validity? |
Multiple treatment inference, reactive arrangements, experimenter effects, pretest sensitisation |
|
How do you increase internal validity? |
Randomly select participants Randomly assign to groups Use a control group |
|
How do you increase external validity? |
Careful adherence to experimental practices More about the research designer Validation studies |
|
What is matching? |
Ensuring that participants in each group are equivalent Expensive Time-consuming May not be possible |
|
What are homogenous groups? |
Groups with members alike on critical factors |
|
What is elimination? |
Eliminating a variable completely |
|
What is constancy? |
Presenting a uniform condition that all participants can experience |
|
What is balancing? What is counterbalancing? |
Balancing: The extraneous variable being experienced in the same manner across groups Counterbalancing: Presenting conditions in different sequences |
|
What is analysis of covariance? |
Equalises differences that might exist between groups and on a covariate |
|
Why must there be a balance of internal and external validity? |
Too much control, little generalisation Too little control, can't make causal statements |
|
What are the advantages/disadvantages of the true experimental design? |
Accurate measurement of variables Control Compare and analyse data Easy to replicate Variables may be narrowly defined Lack of generalisation beyond experimental setting Artifical nature could lead to artificial findings |
|
What is a quasi-experiment and what are the types? |
Same as true experiment except that random assignment isn't possible Non-Equivalent Control Group Design: Participants non-randomly assigned > Pre-test > No/treatment > Post-test Static Group Comparison: Participants assigned > No/treatment > Post-test Single Subject: A Measure behaviour > B treatment > A remove treatment > B Re-add Multiple baseline: Same as single subject but with multiple participants Cross-sectional: Data collected from different types of people at one time Longitudinal: Same group of people observed at different times |
|
Why is random assignment of participants to groups important? |
Ensures that changes in the DV are not due to biased assignment of subjects to experimental conditions |
|
How do you prepare for data collection? |
Construct a data collection form Establish a coding strategy Collecting the data Entering data onto the collection form |
|
How is a data collection table presented? |
One column for each variable One row for each participant |
|
How is coding data presented? |
Using single digits when possible Use simple and unambiguous codes Use explicit and discrete codes |
|
What are the 10 commandments of data collection? |
Get permission from institution review board Think about type of data you will collect Think about where data will come from Be sure the data collection form is clear and easy to use Make a duplicate of the original data and keep in a separate location Ensure that those collecting data are trained Schedule your data collection efforts Cultivate sources for finding participants Follow up on participants that you originally missed Don't throw away original data |
|
What is central tendency? |
"Middle" of data Worked out using mean, mode and median |
|
What is a mode? |
The score/value that occurs most frequently in the distribution Nominal E.g. eye colour |
|
What is a median? |
The score in the middle of the distribution (n + 1)/2 Ordinal E.g. class rank |
|
What is a mean? |
The average Sum of all scores/number of scores Interval/ratio E.g. age in years |
|
What is the equation for outcome? |
Outcome = model + error |
|
What is dispersion? |
The variability or spread of scores in the data E.g. Range, interquartile range, sum of squared errors, variance, standard deviation |
|
What is the range? |
Taking the largest score and subtracting the smallest |
|
What is interquartile range? |
Dividing a set into four equal sections of three values 3rd quartile - 1st quartile = interquartile range |
|
What is deviance from the mean? |
The difference between the mean and a data point |
|
How do you get the total error rate? |
Add up all the deviances in a data set |
|
What is the sum of squared errors? |
The variance of the measured data from the true mean of the data. Uses squares because otherwise scores cancel each other out |
|
What is variance? |
The average error between the mean and the observations made Found by dividing the sum of squared errors by n-1 |
|
What is standard deviation? |
The square root of the variance |
|
What does the standard deviation tell us? |
How well the mean represents the data Small SD (close to mean) = data close to mean Large SD (far from mean) = data distant from mean |
|
What is a frequency distribution? |
A histogram Tallies score occurance X axis = values of observations Y axis = frequency each value occured |
|
What is a grouped frequency distribution? |
Where intervals are counted |
|
What is a normal curve? |
Represents normal distribution Sum of all probabilities for a random variable 68% of scores lie within 1SD of mean 1SD to either side of mean contains 34% of area under curve Mean = median = mode |
|
What is skewness? |
Skewed distribution Positively skewed: Clustered at lower end tail on right Negatively skewed: Clustered at higher end tail on left |
|
What is kurtosis? |
Degree to with tails cluster at ends of distribution Positive/Leptokurtic: Many scores and is pointy Negative/Platykurtic: Light tails and flatter |
|
What are examples of modal characteristics? |
Unimodal: One prominent high point Bimodal: Two prominent high points Multimodal: Multiple prominent high points |
|
How do you calculate probability of certain scores using a normal distribution? |
If the mean is 0 and the SD is 1 a table can be used. If not they must be changed by creating z scores |
|
What do z scores represent and predict? |
Represent scores along the x axis as a percentage of distribution Used to predict the percentage of scores both above and below a particular score Predicts probability that a score will occur in a distribution |
|
Which are the most important z-scores? |
1.96 and -1.96 They represent the top and bottom 2.5% 2.58 and -2.58 They represent 99% of scores -3.29 and 3.29 They represent 99.9% of scores |
|
What is the difference between descriptive statistics and inferential statistics? |
Descriptive stats: Describes essential characteristics (distribution, central tendency, dispersion) Inferential stats: Allow inferences to a larger population from the sample |
|
How are inferences made? |
Representative samples from two groups are selected Participants are tested Means from each group are compared Researchers conclude that differences are from chance or reflect true differences Conclusion is drawn regarding the role group membership plays in observed differences |
|
What is always the first explanation for observed differences? |
Chance Unexplained variability |
|
What is the goal of science? |
To control sources of variability, reducing the role of chance as an explanation |
|
What is the law of large numbers? |
Collecting more experimental data means that actual results get closer to the theoretical expected results |
|
What is central limit theorem? |
Important in generalising Repeated sampling will result in scores that represent the population Samples drawn from a population will be normally distributed |
|
Why will there always be sampling error? |
Sampling is imperfect because you can never select a sample exactly like those in the population |
|
What is the calculation of inferential statistics based on? |
The extent to which the sample from which data has been gathered is a representative of the population |
|
What is the most regularly used method for testing research questions with statistics? |
Null hypothesis significance testing Hybrid of Fisher/Newman & Pearson's theories Tells us whether the alternative hypothesis is likely to be true Calculate p value If p is small, the model fits the data well and concludes support for alternative hypothesis |
|
Who developed the idea of a null/alternative hypothesis? What does it state? |
Neyman and Pearson (1933) States that an affect is absent |
|
Why do we need to try rejecting the null hypothesis? |
Because we can't prove a hypothesis with statistics |
|
What is the difference between a directional hypothesis and a non-directional hypothesis? |
Directional: Predicts the direction of difference E.g. People will be less accurate at matching unfamiliar faces than familiar faces Non-Directional: Does not predict direction of difference E.g. There will be a difference between the accuracy of matching unfamiliar faces compared to familiar faces |
|
How do we test if a model fits the data? |
Comparing systematic variation to unsystematic observation Characterised by a unique distribution of values that are used to evaluate data |
|
How unlikely does a result have to be before deciding that it is significant? |
Alpha (a) Corresponds to p value < 0.05 |
|
How do we test for significance? |
State the null hypothesis Establish significance level Select aproppriate test statistic Compute test statistic Determine value needed to reject null (critical value) Compare obtained value to critical value If value > critical value, reject null If value < critical value, accept null |
|
Which test is performed if there's a difference between the means of two unrelated groups? |
t-test for independent means
|
|
Which test is used if there's a difference between the means of two related groups? |
t-test for dependent means |
|
Which test is used if there's an overall difference between the means of three groups? |
ANOVA Analysis of variance |
|
What are Type I/Type II errors? |
Type I: When we believe there is an effect in a population but there isn't Type II: When we believe there is no effect in the population but there is |
|
What are some issues with null hypothesis significance testing? |
Reliance on merely refuting the null All or nothing thinking P values don't tell us importance only probability |