PSYC 303 Exam 2 Notes
PSYC 303 Exam 2 Notes Psyc 303
Popular in Research Methods in Psychology
Popular in Psychlogy
This 17 page Study Guide was uploaded by Miranda Bostad on Tuesday March 8, 2016. The Study Guide belongs to Psyc 303 at University of North Dakota taught by Dr. Adam Derenne in Spring 2016. Since its upload, it has received 24 views. For similar materials see Research Methods in Psychology in Psychlogy at University of North Dakota.
Reviews for PSYC 303 Exam 2 Notes
Almost no time left on the clock and my grade on the line. Where else would I go? Miranda has the best notes period!
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 03/08/16
Chapter 4 (Lectures 1 & 5) Conceptualization & Measurement Theoretical Construct: An abstraction or concept used to describe or explain a set of empirical observations. o Five factor Theory: AgreeablenessConscientiousnessExtraversionOpennessE motional Stability Are there differences across persons… groups… cultures? Do different tests produce different conclusions? Are there differences within individuals across tests/over time? Measuring and testing theoretical constructs requires operational definitions. Decisions also have to be made on what kind of data to collect, how the data will be collected, and how the data will be aggregated. o “Three Box” Model of Memory (Book for flow chart) Are there differences across persons… groups… cultures? Rate of decline/age Do different tests produce different conclusions? Are there differences within individuals across tests/over time? Not all theories have empirically-grounded constructs, or constructs that lend themselves to testing and measurement. o Psychodynamic Theory (Id, Ego, Superego) Freud Observed for years Based on observations, experience this theory emerged Measurement Scales Nominal Scales: Data are names or categories o What is your gender? Male/Female o How would you describe yourself? Basically introverted/basically extroverted Ordinal Scales: Measures rank order. st nd o 1=1 Place, 2=2 , etc. o Example: List the following activities in order from most preferred to least preferred: Hang out with friends, do homework, etc. Chapter 4 (Lectures 1 & 5) Interval Scales: A scale on which equal distances between scores represent equal differences in the property being measured. o Example: Scores on an IQ test o IQ ranges from 0 on up. o The difference between an IQ score of 90 and 100 is equivalent to the difference between 130 and 140, o An IQ of 0 does not mean that on has no intelligence. o Someone with an IQ of 180 is not 3 times as intelligent as someone with an IQ of 60 o Interval scale numbers cannot be meaningfully multiplied or divided. Ratio Scales: Scores possess all the characteristics of real numbers. o Number of times a patient was hospitalized. o Amount of salivation produced by a dog. o Average hours of sleep per night. o Number of words recalled. Measures of Central Tendency Mean: Average Value Median: Middle Value Mode: Most Frequent Value o Problematic sometimes Which one should be used depends on the data Frequency Distribution: o The “bell curve.” o MEAN = MEDIAN = MODE o A skewed distribution. All values differ. Outliers have huge effect What should researchers do in the following examples? o Age at first birth of Tanzanian women. Chart is not symmetrical (but pretty close) USE THE MEAN o Symptoms of depression among hospitalized older adults. Skewed results Bar graph is higher on left side (Substantial variation) USE THE MEDIAN When normal to skewed you use the median usually o Frequency of problem behavior in children before and after a reprimand. Skewed results After reprimand problem behavior falls Chapter 4 (Lectures 1 & 5) Behavior does not necessarily have a measurement (turn labels into numbers?) Reliability & Validity Reliability: A measure yields consistent scores. o Types: Test-retest, inter-observer reliability, inter-item reliability, split- half reliability. o Observational Research: How much confidence can we have in the observations? Were the operational definitions adequate? Did the observer adhere to the plan? Inter-Observer Reliability: o Do the data from one observer match those of a second observer? Survey Research: o How much confidence can we have in the data? o Body Image Survey Example Designed to assess coping mechanisms: Appearance Fixing Positive Rational Acceptance Avoidance o Do the items on the survey measure what they are supposed to measure? o Did the participants take the task seriously? o Were the participants’ responses affected by their current mood or level of mental awareness? Inter-Item Reliability: o Are the responses consistent among related items? o If yes, then the different items are measuring the same thing? o Appearance Fixing: If participants consistently answer one of these items differently than the others, then the responses from that item are of questionable value. Split-Half Reliability: o Are responses to items in the first half of the test consistent with those in the second half? If results differ, other problematic factors are present. Test-Retest Reliability: Chapter 4 (Lectures 1 & 5) o Are responses consistent when individuals retake the test? Validity: Validity: Degree to which observations or methods are sound and reveal “the truth.” Face Validity: Measurement procedure appears to measure what it is supposed to measure. o “Does this make sense?” o Ex: Do these look like appropriate questions to ask about how people respond to body image threats? Content Validity: The test items fully measure the construct. (True Understanding) o Do these items fully capture what it means to have positive rational acceptance about body image? Theoretical Construct: An abstraction or concept that explains or predicts a set of empirical findings. o Thoughts, attitudes, feelings Construct Validity: The measure actually measures the construct it’s supposed to measure. o Are we actually measuring positive rational acceptance with these items and not something else? o The mental rotation task is one of the tools used in research on cognitive aging. Can we be sure that the task measures the speed of mental processing and not motivation/interest in the task, or familiarity comfort with computerized tasks? o Participants in the antidepressant study are periodically asked questions about their current state. Can we trust the data to accurately tell us about participants’ depression symptom? Convergent Validity: Different ways of measuring a given phenomenon should produce similar results. o Performances on the mental rotation task should correspond with performances on related tasks, self-reports, and natural behavior. Puzzles vs. Mental rotation task o Responses on the depression symptom questionnaire should correlate with other self-report measures, physiological measures, days hospitalized for depression, evaluation by a therapist. Chapter 4 (Lectures 1 & 5) Divergent Validity: If the method is used to measure a different phenomenon, it produces different results. o Mental Rotation Task vs. Mathematical Questions (Computer and age) o Ask questions about depression and other disorders. Things that should be different are different. Chapter 5 (Lectures 2 & 6) Survey Research: Research that involves obtaining self-report data from a large number of people, preferably via random sampling. Measures phenomena that cannot be observed; is practical to conduct. Examples: o What is the incidence of mental disorders in the general population? o What are college students’ attitudes towards sex and dating? o How is procrastination related to conscientiousness and other components of personality? Issues: people might not be willing to give accurate picture about some of these topics (e.g., mental illness, sexual relations) ASD and Folic Acid Example (Correlational Research) o Can we be sure that the sample is representative of the population of interest? Ex: Study participation o How much confidence can we place in the survey instruments/questionnaires? Ex: Measure of ASD o How much confidence can we place in accuracy/confidence of participants? Ex: Measures of Folic Acid Use & Dietary Folic Intake o Is it possible that a third variable accounts for the correlation between folic acid intake and autism? Ex: Whether mother takes supplements may be correlated with education/income—Leads to other factors that may affect ASD Two Basic Concerns: o Can we trust the data? o Can we generalize with confidence? Can We Generalize with Confidence? Researchers typically use a sample of participants drawn from the population, because the population is usually too large. Major concern in survey research: o Do we have a representative sample? That is, a sample that “looks like” the population. o How does the sample compare to the population of Norwegian mothers and children? Or the population of all mothers and children? Possible methods of obtaining participants: Chapter 5 (Lectures 2 & 6) o Probability Sampling: There is a predetermined chance that a given person will be a participant. o Nonprobability Sampling: Selection of participants is not based on chance. Probability Sampling Examples: o Simple Random Sampling: (Computer randomly pulls names from database). Random student that took ACT is surveyed o Systematic Ranthm Sampling: (Every nth person is selected). Every 10 person that walks by o Cluster Sampling: (e.g., random villages). War in Syria, survey random places (towns/community) with different religions/ethnicities (no database, too many people, difficult to travel) Nonprobability Sampling Examples: o Availability Sampling: (Those readily available) Easy access to UND students (not likely to lead to representative sample) o Quota Sampling: (Have advanced idea of who you want your sampling to be) 50 men, 50 women, that’s it. Sample that has certain characteristics that researcher values. Population and Sample Examples: A research firm is hired to learn more about shoppers’ interests at Columbia Mall. The firm interviews 50 shoppers that are in the mall over the course of a Saturday morning. o Will this method produce a representative sample? NO Time frame is problematic/Possibly older population Not all will stop to answer surveyors questions A researcher in the psychology department runs an ad in the Dakota Student offering $20 to students that participate in a 3-hour experiment on learning and memory. o Will this method produce a representative sample? NO Time constraints/Not a lot of money for 3 hours The Grand Forks Herald randomly pulls names out of North Dakota phone books and the calls those people to see how they feel about some proposed legislation. o Will this method produce a representative sample? NO Many people not in phone book More older people in phone book Chapter 5 (Lectures 2 & 6) Population and Sample Volunteer problem: The people that volunteer to participate in research may differ from other people. Survey Research (Cont.): Possible methods of collecting data: o Face-to-face interview o Telephone interview o Mail questionnaire o Internet research Face-to Face Interview o Advantages: People less likely to refuse participation. Possible to judge whether participant understands question and can provide clarification. Easier to ask follow-up questions, if answer judged incomplete. May be possible to tell whether participant is being honest. o Disadvantages: Personal characteristics of the participant/interviewer may affect the responses. The interviewer may bias the results through body language. Participants may not be comfortable being fully honest (reactivity) Face-to-face interviews are expensive and time-consuming. Telephone Interview Mail Questionnaire Internet Research o Cheapest, least time-consuming method. o Similar limitations with mail questionnaires. o Difficult to locate participants; rate of refusal is high. Survey Research (Cont.): Previously: o Differing sampling procedures o Differing modes of data collection (and their pros & cons) Today: o Differing questionnaire formats o Threats to validity in survey research Developing the Survey: Chapter 5 (Lectures 2 & 6) What are some problems with the following questions? o “What kinds of drugs do you take?” May be interpreted in different ways (drug/illicit?) Not specific enough o “How do you feel about your mother?” What if they don’t have a mother Not everyone has the same experiences Questions should be specific and precise. Questions should not make unwarranted assumptions about the respondents. Fixed-Alternative Questions: o Participants forced to given one of several possible answers o “How old are you?” A) 18 B) 18-24 C) 25-30 D) 30+ o Likert Scale: Consists of a series of items pertaining to a topic. The ratings for each item can vary from favorable to unfavorable. The participants’ overall attitude is inferred from the sum of all ratings. Example: Pempek et al. (2009) used a 31-question Likert scale to assess college students’ social networking experiences on Facebook. During the past week how much time have you spent… On Facebook (etc) o Not much o Some o Quite a bit o A whole lot Open-Ended Questions o Participants free to provide a range of answers. “How many siblings do you have?” “How would you describe yourself?” o Allows for more flexible and complete responses. o Responses are more difficult to score. o Example: “Predicting Marital Stability and Divorce in Newlywed Couples” o Content Analysis: Chapter 5 (Lectures 2 & 6) Procedure used to turn written or spoken responses into quantifiable data. Talking about their dog… Sue: Sweetie! She’s not smelly. Bill: Did you smell her today? … Bill: No, you’d better be careful. Sue: No, you’d better be careful… Don’t call my dog oily, boy. Strategy 1: Count the number of each type of utterance (statement). I hate that he doesn’t make time for me I one negative utterance. Strategy 2: Classify each type of utterance. One possible approach: responses are classified as “disinterest,” “refutation,” “acknowledgement,” or “validation.” Another possibility: responses are rated on a scale from 1 “disagreement” to 5 “agreement.” Developing the Survey (Cont.): What factors undermine the validity of survey responses? Social Desirability: Participants distort their responses in a manner that portray themselves in a positive light. o Circle all the traits that describe you: Caring, Moody, Generous, Warm, Difficult, Tired, etc.… More likely to pick caring, warm, generous… o Do you believe that all eligible Americans should vote? o Do you believe that it is important to always obey the law? o Do you believe that America should recycle its natural resources? All most people would say yes to, but doesn’t mean they do all the time. Social desirability causes them to answer in a certain way, although it might not correlate with their actions. o Motivated Forgetting: A tendency for past events to be recalled in a distorted form. Events are remembered in a manner that is self- serving. How many alcoholic drinks do you have during an average week? A) 0-1 B) 2-4 C) 5-7 D 8-10 Context Effects: The context in which a question is asked (i.e., the information that precedes it) can affect participants’ responses. Chapter 5 (Lectures 2 & 6) o 1. How happy are you with life in general? o 2. How happy are you with your dating life? More favorable response to second question in this order vs. reversed order. Chapter 6 (Lectures 3 & 4) Example of Experiments: We want to examine the effectiveness of a new antidepressant. o What do we need to do to conclusively determine whether or not the drug works? o Complications: (Potential Confounds) People w/ depression vary in terms of which symptoms they manifest, how severe those symptoms are, how long the symptoms have been present, the factors that precipitated depression, and whether or how those symptoms were previously treated. Some people naturally improve on their own Expectations can influence outcomes (e.g., the placebo effect). Drugs that are generally effective do not necessarily produce improvement in every user. o Problem: Some people naturally improve on their own/others don’t benefit even when the drug works. o Solution: There is probably no way to identify these people in advance. Through random assignment, equal proportions should end up in both groups. Works best with a large sample. o Problem: Expectations can influence outcomes (e.g., the placebo effect). o Solutions: Can’t eliminate expectations. It’s not a problem so long as expectations are equivalent within each group. The True Experiment: 1.) There are at least 2 levels of the independent variable. o (There must be a comparison; it doesn’t have to be with a control group) 2.) The design controls for threats to internal validity. o (No confounds) 3.) Participants are randomly assigned to conditions. o (Randomization controls for potential confounds both known and unknown) Hypotheses: Null Hypothesis: The groups are not different and the independent variable does not affect behavior. Chapter 6 (Lectures 3 & 4) Confound Hypothesis: The groups are different, but not because of the independent variable. Research Hypothesis: The groups are different because of the independent variable. Hypothesis Examples: o Null Hypothesis The antidepressant has no effect. o Confound Hypothesis The antidepressant falsely appears to have been effective. o Research Hypothesis The antidepressant is effective. The researchers need to convince themselves and their peers that the research hypothesis is correct and the confound hypothesis is not. o (Where peer review is used) Examples Cont: o Null Hypothesis There is no reason to believe Robert was the murderer. o Confound Hypothesis There is reason to believe Robert was the murdered, but it was actually someone else. o Research Hypothesis We believe Robert was the murderer. The detectives and prosecutors have to convince themselves and a just that the “research hypothesis” is correct and that the “cofound hypothesis” is not. Examining Validity: How were participants selected and assigned to groups? Are the groups different in some nontrivial way? Aside from the IV, did the groups have different experiences? Was the performance of the different groups measured in the same way? Wittlinger et al. (2006) o How do desert ants navigate? o Hypothesis: Ants count the number of steps to/from the colony. o Group 1: No Change o Group 2: Legs shortened o Group 3: Ants given stilts o Examining Validity: Selection: Easier to modify slower ants? Experiences: Pain? Uncomfortable? Human Example Chapter 6 (Lectures 3 & 4) o A researcher wants to determine whether relatively young and relatively old adults differ in their ability to perform a mental rotation task. o The younger adults are college students working for extra credit. The older adults are retirees recruited through a TV ad and offered $20. What threat to internal validity stems from this feature of the method? No control of education/intellectual differences/motivational differences/Lack of technology knowledge in older? Assignment Procedures: A study examines whether participants that take a new drug show fewer symptoms of depression that participants taking a placebo. People with depression vary in terms of which symptoms they manifest, how severe those symptoms are, how long the symptoms have been present, the factors that precipitated depression, and whether or how those symptoms were previously treated. Participants: o Female, 26, mod., Male, 55, mild, Female, 39, mod., Female, 31, mild, Male, 32, severe, Female 53, severe…… Assignment is random. o Sample ½ in control ½ in treatment o Group 1: & Group 2 Similar mixture of characteristics CONCERNS: o What if individual differences are not evenly distributed? SOLUTIONS: o Elimination procedure: Participants with certain characteristics are excluded from the study. Limiting gender? Only female Limit age Limit severity of symptoms MAKE IT EQUAL o Matching (Equating) Procedure: Participants are first matched into homogenous blocks and then randomly assigned. Boys in Sample 2 Female/2 Male Possible matched variables in the drug study: Gender Severity of depression Chapter 6 (Lectures 3 & 4) Nature of depression (state vs. trait) Concurrent use of psychotherapy Limitation: With multiple potentially important sources of chance variation, perfect matching may not be possible. Within-subjects design: A single group of participants is given different levels of the IV at different times. Between vs. Within: o Imagine that you are interested in the effects of two different types of background music (classical music and rock music) on people’s ability to remember what they learn. Subjects are given 12 min to study a list of words, then they have 3 min to write down as many of the words they can remember. o Between-Subjects Design: Chief Disadvantage: A difference in the groups rather that the IV? o Within-Subjects design: Chief Advantage: Subjects act as their own controls Eliminates concern over whether individual differences are even distributed. Disadvantages of Within-Subjects designs: 1.) Time of measurement problem: The time at which a condition occurs is confounded with the independent variable. o Participants’ alertness or mood may change with time. 2.) Carryover Effects: The effect of one condition may “contaminate” subsequent performances. o The music may have long-lasting effects on alertness or mood. 3.) Subject Attrition: Participants may drop out during a long study. Participants to leave/ stay may be atypical. o Participants with extreme scores may drop out before the study is complete. Chapter 6 (Lectures 3 & 4) 4.) Order Effects: Repetition of the task affects performances per se, either because subjects become fatigued, or because of practice effects. o Subjects are better practiced at Time 2 than at Time 1. The Essential Control Procedure: Counterbalancing: Different participants are given the conditions in different orders. o ½ of participants: Classical, then rock. o ½ of participants: Rock, then classical. What else can the researcher do to mitigate… Carryover effects? o Put time between/Another Day Time of measurement? o Same time of day (10 A.M. & 10 A.M.) Subject Attrition? o Reward Structure/Strengthening o Increasing money with sessions Order Effects? o Spread things apart in time (for fatigue) o Make things as different as you can(for practice effects) Counterbalancing (cont.) Imagine that you are interested in the effects of four different types of background music. (Classical, rock, country, and hip- hop)…Same as above. Simple Counterbalancing: o Classical; Rock; Country: Hip Hop o Rock; Country; Hip Hop; Classical o Country; Hip Hop; Classical; Rock o Hip Hop; Classical; Rock; Country o Why is this not an optimal arrangement? Always the same ones coming after each other Ex: Classical puts people to sleepRock always follows classical. Complete Counterbalancing: o Super long list of lists Chapter 6 (Lectures 3 & 4) o With more than a few conditions, the number of possible sequences becomes unmanageable o Levels of the IV Possible 2 2 3 4 24 o Each condition must occur equally often. o Each condition must occur at each possible point in time o Each condition must precede and follow all other conditions equally often.
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'