Research Methods, Final Exam Study Guide
Research Methods, Final Exam Study Guide Psych 305
Popular in Research Methods
Popular in Psychlogy
This 19 page Study Guide was uploaded by Clarissa Hinshaw on Thursday May 5, 2016. The Study Guide belongs to Psych 305 at Northern Illinois University taught by Keith Millis in Winter 2016. Since its upload, it has received 59 views. For similar materials see Research Methods in Psychlogy at Northern Illinois University.
Reviews for Research Methods, Final Exam Study Guide
I love that I can count on (Clarissa for top notch notes! Especially around test time...
-Nathanael Gerhold Sr.
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: 05/05/16
Chapter 1: Scientific Understanding of Behavior Intuition: making predictions based on a gut feeling. Example: You and your significant other are usually very close and open with each other. One day they do not answer your phone calls or return any of your texts when they normally do so. You have a gut feeling your significant other is cheating on you even though you have no evidence to support this claim. Scientific research: finding answers to scientific questions based on experiments, correlational studies, case studies, surveys, longitudinal studies and other forms of research. Example: A longitudinal study could be done to see if samesex parents adopting children is correlated with homosexual or bisexual orientations in their children. To conduct this study, a sample of same sex parents and children could be selected. Twenty years later, researchers would follow up with these families to see what the children’s sexual orientations turned out to be as adults. Limits of intuition o Reliance on beliefs without research o Biased o Can cause illusionary correlations. Taking the examples from the significant other not texting you back and your belief of cheating. You are not taking into account other causes for your significant other’s behavior such as being busy, having a bad day, or just needing some alone time. Limits of authority o Reliance for accurate information. o Sometimes lack of scientific evidence needed to form correlations. Qualities of good scientific research o Empiricism: Research collected through observation and data collection. o Falsifiability: The ability to form a hypothesis (scientific prediction) and possibility for the hypothesis to be rejected. If the hypothesis cannot be rejected, it is not scientific research. o Peer review: research critiqued by other scientists in the field of study to assure accurate information is published. Pseudoscience includes o An untestable hypothesis o Questionable data o Unsited references o No conflicting evidence o Vague, unrevised information Basic Research: Used to find out information about the nature of behavior. The results are not intended to create change, but are sometimes used in government policies. Example: conducting a study on marriage attitudes among children of divorced couples. This study would not be intended to change these attitudes. Applied Research: Research intended to create social change. Example: A study to see if high school abstinenceonly programs increase or decrease rates of adolescent pregnancy. The results of this study could help decide whether abstinenceonly programs continue in school districts. Chapter 2: Where to Start Hypothesis: a statement of whether one variable is correlated with or caused by another. This is proven or disproven through research studies. Example: studying by selftesting, increased exam scores. Prediction: a belief or disbelief of the hypothesis will be proven or disproven from results of the study. Example: I believe the hypothesis regarding studying and test scores will be proven. Participants: people who agree to take part in a research study. Other terms include respondents, subjects, and informants, depending on the type of study. Where do ideas come from? Common sense: ideas people believe to be true and assume are universal. Examples: Do opposites attract? Do birds of a feather flock together? Do actions speak louder than words? Does everything happen for a reason? Observation: developing project interests based on life experiences. Example: a person doing a study of the effects of autism on socialization based on their own experiences. Serendipity: ideas discovered by accident or luck. Theory: possible explanations for different types of behavior. Past Research: Used to guide new research studies and write literature reviews. Library Research Methods Journals: Places where researchers publish their results. They are reviewed by scientists and published as legitimate journal sources (peer review). Psychological Abstracts: Summaries of research articles published each month, found through databases. Example: PsychInfo. Search can be done by typing in subject of interest. Linking two topics with the words AND, OR, or NOT can narrow or widen results. Most contain links to full text. Ones that don’t can be requested from other libraries. There are also setting to see only peerreviewed articles and most recent articles first. Other databases include Science Citation Index and Social Sciences Citation Index. Literature Reviews: articles summarizing research from other studies. Internet (basic Google Search): Though this method often contains results from any topic, websites must be clearly evaluated for legitimacy. Google Scholar: A search engine for scholarly journals. Good for some research, but many results don’t contain full text articles for free any articles aren’t peer reviewed or as scholarly as database articles. Professional meeting searches: contains articles presented at professional conferences. Parts of a research article Abstract: overview of the topic Introduction: Stating research topic and hypothesis Method: states how the hypothesis was tested Results: the outcome of the hypothesis Discussion: What the results mean for science and the world. Ethics Beneficence: maximizing benefit and minimizing risk when conducting research. Physical harm: negatively affecting the health of a participant. Example: giving a pregnant woman alcohol to examine the effects on her future child. Stress example: a participant may feel psychological distress if they are taking a survey about suicide and the survey prompts old memories of a past suicide attempt. Confidentiality: information from studies must be kept private. Example: studies analyzing the past behavior of addicts. Informed consent: participants must be told about what they will be doing in the study, the purpose, and their right to withdraw at any time. They must sign an informed consent form. Autonomy issues: Lack of autonomy: some populations are not able to give consent on their own. Examples: children, people with certain disabilities, or people under the influence. Coercion: guilttripping people to participate. This includes participating to avoid an unpleasant consequence (taking a survey to fail), or to gain a reward (participating in a study to earn a bonus at work). Think positive and negative reinforcement! Withholding information and deception: sometimes participants are not told everything about a study to keep unbiased results. Example: seeing student essay patterns when told the essays for career services. This is usually okay as long the deception does not cause harm and is explained at the end of the study. College studies usually do not mind studies with deception. Debriefing: telling participants the true nature of the study after completed. They can then learn about what the study is testing for as well as predicted results. Deception alternatives Roleplaying: participants are asked how they would react in a certain situation. Example: How would you react if your significant other cheated on you? If you saw someone homeless on the street? If Donald Trump won the presidential election? These reactions are all hypothetical and could change if they actually happened or if the situation changed. Simulation Studies: used to reenact realworld situations. Sometimes goes too far, such as in the Stanford Prison Experiments. Justice: giving benefits and minimizing risks to participants. Institutional Review Board (IRB): reviews all studies involving human subjects to make sure they are ethical and follow those guidelines. Exempt research: studies with no risk. Can include observations, anonymous surveys and using research from previous public data. Minimal risk: risk is low to participants. Example: reading about frogs and testing curiosity. Greater than minimal risk: high physical or psychological risk studies. Example: a study about trauma from rape or sexual abuse. Ethics for animal research: although animal research is declining, it is still controversial, especially among animal rights groups. Here are ethical guidelines to make sure any animals used are being treated as kindly as possible. Fabrication (publishing false results) and plagiarism (reporting someone else’s work as you own or not using proper citation) is unacceptable. Chapter 4: Studying behavior Variables Situational Variables: variables describing a particular environment Response Variables: Behaviors or responses of people Participant or Subject Variables: variables reflecting differences among people Mediating Variables: variables measuring situational variables. Diffusion of responsibility Helps explain mediating variable Receives from number of bystanders and gives to helping behavior Operational definitions of variable: the researcher first has to come up with ways to study the behavior in question and figure out ways to manipulate a variable. This variable is defined on how it is measured. This strategy helps researchers to think critically and outside of the box. It sometimes doesn’t make ideas specific enough. Relationships between variables: the variables are being tested to see if there is a relationship between them. Some variables are categorical, while others are numerical. Positive linear relationship: a positive correlation. Example: there is a positive relationship between acceptance of homosexuality and the amount of gay people feeling safe enough to come out. Negative linear relationship: embodies a negative correlation. Example: there is a relationship showing the increase in number of hours people spend on social media and decrease in exam scores. Curvilinear relationship: shows a positive/negative relationship changing over time. Example: there is a positive relationship between fad dieting and weight loss at first, but a negative relationship between these variables later on. No relationship: Variable with little or no correlation. Example: there is no relationship between a cisgender woman’s breast size and amount of milk she can produce for a baby. Correlation Coefficient: a number between 1&1, showing how strong or weak a correlation is. The closer to 1 or 1, the stronger the relationship. The closer to zero, the weaker the relationship. Positive numbers=positive correlations, and negative numbers =negative relationships. Random variability and error variance refer to the uncertainty and randomness of events. Experimental method: research method where subjects are in labcontrolled environment. NonExperimental method: a method where variables are not controlled, but correlational relations are examined. Also referred to as the correlational method. Directionality issue: is one variable causing the other in a correlation, or is it the other way around? Example: Do bad grades cause anxiety, or does anxiety cause bad grades? Third variable problem: are there other variables possibly causing both scenarios. Variables which could cause both bad grades and depression include social media, unhealthy personal relationships, lack of financial stability, and working too many hours during the school year. Experimental method: studies taken place in a controlled environment with at least 2 groups. The experimental group(s) receives manipulation, while the control group does not or receives a placebo (fake treatment). Example: an experiment testing the effects of anxiety pills would give the pill to the experimental group and a sugar (placebo) pill to the control group. Experimental control: making sure the variable tested is the only one manipulated and all other variables remain the same for all groups. Randomization: randomly assigning people to groups to eliminate bias. The independent variable causes the dependent variable. Example: in a study measuring whether listening to a certain type of music influences personality, the music would be the independent variable and personality would be the dependent variable. Artificiality of experiments: sometimes the lab can create an artificial atmosphere. Field experiment: where a variable is manipulated in a natural setting. Example: a cashier being constantly told to ask a customer for more personal information and to try and sell more products to see how well the cashier conforms to orders. Sometimes it is unethical or impractical to use lab or field experiments, so correlational methods must be used. These can be analyzed by ex post facto: after the fact. Participant variables: demographic information of participants. This can include age, gender, race, birth order, personality, or marital status. Types of validities: Construct validity: is the experiment following the operational definition? Internal validity: ability to assess results for relationships. External validity: ability to draw conclusions from results to other populations. Conclusion validity: being able to analyze data for accurate relationships. Also referred to as statistical conclusion validity. Chapter 5: Measurement Concepts Reliability: the stability of a behavioral measure. Should gain similar results with different trials. Components of reliability measure True score: real score of a variable Measurement of error: possible area of measure, which could be off. TestRetest reliability: testing a variable twice and comparing scores. Internal consistency reliability: Splithalf reliability: splitting both correlation scores in half. Cronbach’s alpha: combining each item on a test to form a correlation. Interrater reliability: how much raters agree on observations. Reactivity: if a person is aware they are being measured, their behavior will be altered. Different from how they would behave in a natural setting. Types of measurement: Nominal: non numeric, also called categorical. Ex: gender Ordinal: ranking. Ex: restaurant or movie ratings Interval: literal measure scores. Ex: intelligence, temperature Ratio: comparison above 0. 0 means variable doesn’t exist. Ex: weight, age. Chapter 6: Observational methods Quantitative approach: includes larger samples not observed as closely. Example: observing people at the mall. Qualitative approach: smaller samples observed closely. Example: case studies, interviewing family members for a paper. Naturalistic observation: also called field work or field observation; making observations in a particular natural setting. This could happen at schools, workplaces, bars, etc. Problems with naturalistic observation: Participation and concealment: Researchers need to decide whether or not to act like a participant, and whether or not to disclose all information. Observers must act like outsiders in a naturalistic observation. Seeing familiar faces can mess with the data. If people know they are being observed they are influenced and act differently. Sometimes participants forget they are being observed. Ex: reality stars forgetting cameras are in their house and revealing personal information. This can cause ethical concerns. o Not always convenient. Ex: there may not be many people in a store to observe. o Doesn’t always explain an event in a way the researcher wants it to. Systematic Observation: observing specific behaviors in one place. Ex: watching children play together on the playground. Coding system: used to measure behaviors. Resident independent behavior: doing something independently. Ex: grooming, sleeping. Resident dependent behavior: asking for help on a task. Staff independencesupporting behavior: reward for being independent. Ex: good grade for writing an essay by yourself, rather than copying off a classmate. Staff dependencysupportive behavior: helping or encouraging other to seek help. Ex: encouraging a depressed person to seek counseling. Other unrelated behaviors. Methodological issues: Equipment: some methods still use a human observer, but most studies involve a camera. Reactivity: knowing the experimenter is there can contribute to altered pattern of behavior. Reliability: used to see if study results are stable. Sampling: data taken longitudinally are usually more accurate than shortdata. Case study: research using a few particular individuals. Ex: interviewing family members about their attitudes toward relationships. Psychobiography: using psych to explain a person’s life. Ex: a famous historical psychologist having a specific way of practicing therapy because of the way they were raised. Archival research: using existing data rather than collecting new data. Ex: analyzing public records or information already in a computer system. Survey archives: data from surveys stored online for people to use. Written and Mass Communication Records: diaries letters and others showing researchers past behaviors. Content analysis of documents: analysis of existing documents. Requires using coding systems. Ex: using this to address marriage licenses. Chapter 7: Survey Research Response set: responding to all survey questions from one perspective. Closed ended questions: give limited response options. Ex: multiple choice. Open ended questions: give more choice for response. Ex: short answer or free response questions. Types of surveys Questionnaires: questions are in written format and participants write their answers. Advantages: cheaper, anonymous Disadvantages: if the participants cannot read or understand the questions, some may not find it boring. Interviews: done face to face to face or over the phone, Skype, Facetime, etc. Interview bias: showing preference or expectation of some participants over others. Ex: showing more bias toward friends or family than strangers or acquaintances. Panel study: a study with more than one person at a time. Confidence interval: the amount of certainty a person has about a variable lying within a specific range. Ex: a person can be 95% confident of their test score being a C or better. Stratified random sampling: choosing an amount from each category. Cluster sampling: a number of groups out of a large number of groups. Ex: sampling 2 groups out of 1,000. Haphazard sampling: a sample done out of convenience. Ex: having your friends and peers complete a survey in class. Purposive sampling: sampling from a specific group. Ex: college students, transgender individuals. Quota sampling: data where the sample reflects the percentage in the population. Ex: if Asians make up 10% of the population, 10% of the sample will be Asian. Chapter 8: Experimental Design Confounding variable: a variable which varies along with the indirect variable. Ex: doing an experiment in different rooms. Experimental Group: the group where a variable is manipulated. Ex: a group who takes a healing herb to see if it heals. Control group: the group with no manipulation. Ex: the group receiving a placebo herb. Posttestonly design: consists of gathering participants, introducing the independent variable and measuring its effect on the dependent variable. Selection Differences: the participants cannot be drastically different. Mortality: the people who drop out of an experiment. Independent group design: random assignment where each participates in one group. Repeated measures design: participants measured after getting each variable level. Order effect: order of showing treatments. Counterbalancing: all the possible orders a presentation can come in. Latin square: orders of conditions Matched pairs design: matching people to a condition based on characteristics. Chapter 9: Conducting Experiments Confederate: a participant in the experiment who is part of the manipulation. fMRI: a machine used to measure brain activity. Sensitivity: measures liking. Ceiling effect: when the independent variable has no effect on the dependent variable. Floor effect: when the task is too difficult to do well. Demand characteristics: features of an experiment which informs participants about the purpose of the experiment. Filler items: items to add to a questionnaire. Placebo group: an experimental group which receives a fake sugar pill in tests for different drugs. Experimenter bias or expectancy effects: an experimenter bringing their own opinions into the study. Pilot study: a trial run with a few participants. Manipulation check: a check to see if the manipulation has the effect it’s supposed to have on the participants. Chapter 10: Complex Experimental Designs Factorial design: designs with more than 1 independent variable. Main effect: the most common effect of the independent variable. Interaction: the two independent variable working with one another. IV x PV designs: includes experimental and nonexperimental variables. Moderator variable: this influences how the 2 variables communicate with each other. Simple main effects: looks at differences in each level of an independent variable. Mixed factorial design: the combination of the 2 above methods. Chapter 11: Single Case, QuasiExperimental, and Developmental Research Single Case Experimental design: also called single subject, single case, or single participant. Experiments following one person. Baseline: the control group before the manipulation. Reversal design: contains a baseline period, followed by a treatment period, then another baseline period. All referred to as withdrawal design. Multiple baseline design: when a behavior only changes after a manipulation occurs. Across situations: The same behavior is recorded in different settings. Ex: school, work, home. 5 Steps of program evaluation: needs assessment, program theory assessment, process evaluation, outcome evaluation, efficiency assessment. Quasiexperimental designs: a design showing effects of an independent variable without a control group. Onegroup posttestonly design: an experiment without a control group, a ‘one shot case study’. Onegroup pretestposttest design: measuring participants before and after a manipulation. Ex: seeing if a training program reduces smoking. History effects: confounding variables or other things occurring between the 2 events. Maturation effects: how people change over time. Ex: people not smoking later in life b/c they are more concerned about their health. Instrument decay: occurs when observing gain understanding, become tired, or change their expectations of the experiment. Regression toward the mean: aka statistical regression, when participants have very high or low scores. Nonequivalent control group design: differences between the groups becoming a confounding variable. Selection differences: groups forming from natural groups. Ex: giving one group the training program and the other group no training program. Nonequivalent control group pretestposttest design: giving participants the measure before and after the manipulation. Interrupted time series design: used to observe changes over an extended period of time. Ex: deaths rates from car accidents over time. Control series design: finding a control group to improve the interrupted time series design. Crosssectional method: studying how things change over time by examining people of different ages. Longitudinal method: measuring subjects over time. Ex: observing attachment styles of infants and parents, coming back 20 years later, and measuring these babies’ attachment styles in relationships as adults. Cohort: a generation of people born around the same time. Ex: baby boomers, generation X, millennials. Sequential method: a compromise between the longitudinal and crosssectional methods. Chapter 12: Understanding Research Results: Description Data can be analyzed by comparing group percentages. Ex: examining if race has anything to do with crime. You would investigate the crime rates for each race. Frequency Distribution: the number of people who receive each score on a variable. Pie chart: a chart showing percentages Bar graph: a graph representing each distinct piece of info. Frequency polygon: a graph using a line of best bit. Descriptive Statistics: research describing specific data. Central tendency: explains the overall sample. Mean: average Median: middle number Mode: # occurring most often Variability: the amount of spread in a distribution Standard deviation: deviation of scores from the mean on average. Variance: the square of the standard deviation. Correlation coefficient: a # showing how strong of a relationship two variables have. Pearson productmoment correlation coefficient: used when both variables have scale properties. Effect size: the strength of which 2 variables are related to one another. Regression equation: this is used to predict a person’s variable score. Multiple correlation: many factors contributing to a certain result. Ex: high school grades, ACT score, and letters of recommendation can all predict college success. Third variables can contribute to both variables in a correlation. Ex: time spent on social media can influence both time spent studying and grades received. Partial correlation: a way to help control 3 variables. Path analysis: a diagram showing arrows leading to each variable. Chapter 13: Understanding Research Results: Statistical Inference Inferential Statistics: using the results to predict what would happen if the study was replicated. Null hypothesis: hypothesis stating no relationship between 2 or more variables. Research hypothesis: the hypothesis stating a relationship between 2 or more variables. Statistical significance: the likelihood of the same results occurring in the general population, rather than by chance. Probability: the likelihood of something happening. Degrees of freedom: df, the sample size of a ttest. Ftest: aka analysis of variance, the test used when calculating anova. Systematic variance: deviation from the mean score. Error variance: deviation of individual scores from their group means. Type 1 error: rejecting a true hypothesis. Ex: doing a csection on a mother who doesn’t need it. Type II error: accepting a false hypothesis. Ex: not doing a Csection when it is needed. Power: a sample size possible to reject the null hypothesis. Chapter 14: Generalizing Results Replication: repeating a study done by a previous experimenter. Exact replication: doing a study the exact same way the previous researcher did and expecting the same results. Conceptual Replication: replicating the main themes of the study, but different procedure. Metaanalysis: combining the results of many studies.
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'