PSYC2900U Lecture 7 - Non-Experimental Research PDF

10/21/24 2 Lecture Overview...

10/21/24 2 Lecture Overview u Overview of non-experimental research Ch. 6 – Non-Experimental u Correlational research Research u Complex correlation u Qualitative research u Observational research PSYC 2900U – RESEARCH METHODS KARLA EMENO NOTE: The textbook goes into statistics occasionally for this chapter – You do NOT need to know how to compute stats for this course. 1 1 2 Assignment #2: Literature Review (20%) 3 4 Assignment #2: Literature Review u Due: Submit in Canvas by 11:59 pm on Friday, October 25th u For this assignment, you will write a Literature Review that incorporates a minimum of 8 peer-reviewed psychological papers (ideally from u Writing an effective literature review: Assignment 1, but you can select new ones if you prefer). u https://www.youtube.com/watch?v=Vc_Yu_61Ymg u This literature review will summarize past research, make it clear how your work contributes/adds/builds upon what has already been done, and u Ontario Tech Library Literature Review Guide: mention why this is an important area of study. u https://guides.library.ontariotechu.ca/literaturereview u You will end Assignment 2 with a Current Study section that outlines your research question and hypothesis (or hypotheses). Your variables and the u Focus more on Steps 3 and 4 – You essentially did general study idea should be briefly introduced here as well. Steps 1 and 2 for Assignment #1 u Instructions for Assignment 2 (including the grading rubric), as well as an example template, have been added to Canvas. 3 4 5 When to use non-experimental 6 Non-Experimental Research research? General Situations: Examples: u Non-Experimental Research – Research that lacks the 1. When you only have a single 1. How accurate are people’s first manipulation of an IV. It involves measuring variables as variable impressions? they naturally occur (in the lab or real world). 2. When you have a non-causal 2. Is there a correlation between relationship between verbal intelligence and u Experimental research can draw causal conclusions variables mathematical intelligence? between an IV and DV and non-experimental research 3. When there is a causal 3. Does damage to a person’s cannot. relationship, but you can’t hippocampus impair the formation (practically or ethically) of long-term memory traces? u Non-experimental research is still just as important as randomly assign participants 4. What is it like to be a working experimental research! 4. When you ask a broad or mother diagnosed with exploratory question depression? 5 6 1  10/21/24 Types of Non-Experimental 7 Internal Validity Revisited 8 Research u Recall that an experiment is high in internal validity IF the 1. Correlational Research – The researcher measures 2 way it was conducted supports the conclusion that the IV variables with little or no attempt to control caused any observed differences in the DV. extraneous variables and then assesses the relationship between them. 2. Observational Research – The researcher makes observations of behaviour in a natural or lab setting without manipulating anything. 7 8 Correlational Research 9 Scatterplots 10 (used to present correlations between 2 quantitative variables) u Correlational Research – Research where the statistical relationship between 2 variables is assessed with little or no effort to control for confounds. u Often higher in external validity than experimental research. r =.51 u Used over experimental research when: 1. Not interested in causal relationships 2. IV cannot be manipulated 3. Want to establish the reliability and validity of measurements 9 10 11 12 Pearson’s Correlation Coefficient (r) Types of Correlation When one trait or behavior accompanies another, we say the two are correlated. Indicates strength of relationship (0.00 to 1.00) Correlation Coefficient r = + 0.37 The correlation coefficient (r) is a Indicates direction (i.e., no correlation) statistical measure of the of relationship relationship between two variables. (positive or negative ) 11 12 2  10/21/24 When Pearson’s r can be misleading 13 Nonlinear Relationships 14 1. Pearson’s r is NOT a good measure for nonlinear relationships. uIt is a good measure only for linear relationships, in which the points are best approximated by a straight line. 2. Pearson’s r is misleading when there is restriction of range. u This means that one, or both, of the variables is truncated and does not vary enough to detect a correlation. 13 14 Restrictive Range 15 Correlation is NOT Causation 16 u Correlations NEVER prove cause and effect. Overall r = -.77 Restricted range r = 0 15 16 Correlation Does Not Imply Causation 17 Complex Correlation 18 u Complex Correlational Research – Involves measuring several Directionality variables and assessing the statistical relationships among them. Problem u Often used to assess the validity of new psychological measures. X could cause Y or Y could cause X Third-Variable Problem A third variable (Z) could cause both X and Y 17 18 3  10/21/24 Factor Analysis 19 Regression 20 u Factor Analysis – A complex statistical technique in which u Regression – A statistical technique that allows researchers to researchers study relationships among a large number of predict one variable given another. conceptually similar variables. It organizes the variables into a smaller number of clusters, with each cluster representing one u Predictor variable (X) – The variable(s) used to make the underlying construct. prediction u Ex:Big 5 personality traits are factors (conscientiousness, u Outcome (criterion) variable (Y) – The variable that is being agreeableness, neuroticism, openness, and extraversion) predicted. u Factors are NOT categories – they are constructs that are u Simple Regression – Involves using one variable to predict independent of each other another. u Ex: Someone can be high in both extraversion AND neuroticism u Multiple Regression – Involves using several variables to predict u Researchers must interpret and label the factors an outcome variable. 19 20 21 22 What is qualitative research? Quantitative Research Weaknesses: Collects large amounts Strengths: Generally begins of relatively Describes their data u Does not provide detailed with a less focused “unfiltered” data from u Provides precise using non-statistical descriptions of the behavior research question relatively small answers to specific of particular groups in number of individuals research questions particular situations u Draws general u Does not communicate Examples of qualitative research: conclusions about what it like to be a member u Interviews human behavior of a particular group in a u Focus groups particular situation 21 22 23 24 Qualitative Research Quantitative-Qualitative “Debate” Strengths: u Quantitative and qualitative research do not coexist in u Helps researchers generate Weaknesses: complete harmony. new and interesting research u Potential lack of u However, most now agree that the two approaches can questions and hypotheses objectivity and should be combined into mixed-methods research. u Rich and detailed descriptions u Difficult to evaluate u Can use qual. research for hypothesis generation and of human behavior in real- statistically (reliability ad quant. for hypothesis testing. world contexts validity) u Triangulation – Use both quantitative and qualitative u Can convey what it like to be u Does not allow for a member of a particular methods simultaneously to study the same general generalization group in a particular situation questions and to compare the results. 23 24 4  10/21/24 25 26 Observational Research Naturalistic Observation u Observational Research – Refers to several types of non- u Naturalistic Observation – Involves observing people’s behaviour experimental studies in which behaviour is systematically in the environment in which it typically occurs (ex: observing observed and recorded. shoppers in a grocery store, children on a school playground). u Some examples include: u Disguised Naturalistic Observation – When researchers make their 1. Naturalistic observation observations as unobtrusively as possible so that participants are not aware that they are being studied. 2. Participant observation u Undisguised Naturalistic Observation – When the participants are 3. Structured observation made aware of the researcher presence and monitoring of their 4. Case studies behavior. 5. Archival research u Reactivity – Being observed changes participants’ behaviour. 25 26 27 28 Participant Observation Structured Observation u The researcher makes careful observations of one or more u Participant Observation– Researchers become active specific behaviours in a particular setting that is more structured participants in the group or situation they are studying. than those used in naturalistic or participant observation. u Disguised Participant Observation – The researchers u Emphasis is on gathering quantitative (rather than qualitative) pretend to be members of the social group they are data. observing and conceal their true identity as researchers. u Involves precise specification of the sampling process and u Undisguised Participant Observation – The researchers coding (for observations requiring a judgment). become a part of the group they are studying, and they u Coding – Clearly defining the set of target behaviours so disclose their true identity as researchers to the group. different observers code them in the same way (need to demonstrate inter-rater reliability). 27 28 29 30 Case Studies Archival Research u Case Study – An in-depth, and often longitudinal, u Archival research – Involves analyzing archival data that examination of an individual. have already been collected for some other purpose. u Often the individual has a rare or unusual condition u Measurement can be more or less straightforward in or disorder or has a damage to a specific area of the brain. archival research. u Mostly qualitative in nature. u Might involve content analysis, which requires specifying keyword, phrase, or ideas and then finding all u There are issues with both internal AND external validity. occurrences of them in the data; these occurrences can u Experimenter biases could impact the case description. be counted, timed, or analyzed in a variety of ways. 29 30 5  10/21/24 31 32 Studying Change Over Time u Cross-Sectional Studies – Compare two or more pre-existing Questions groups of people (no manipulation of an IV or random assignment). u Cohort Effect – Differences between the groups may reflect the generation that people come from rather than a direct effect of age about u Longitudinal Studies – One group of people is followed over time as they age. Chapter 6? u Cross-Sequential Studies – People in different age groups are followed over a smaller period of time (combines elements of both cross-sectional and longitudinal studies). 31 32 6