PSYC 2900U - Lecture 3 - Ch. 2 Getting Started in Research

9/16/2024 2 Lecture Overview...

9/16/2024 2 Lecture Overview Model of scientific research in psychology Ch. 2 – Getting Started Finding a research topic in Research Generating good research questions Developing a hypothesis Designing a research study PSYC 2900U – RESEARCH METHODS Analyzing the data KARLA EMENO Drawing conclusions and reporting the results Assignment 1 reminder 1 1 2 3 4 Scientific Method Step 1: Choosing a research question Step 2: Conducting a literature review 1. Formulate a testable hypothesis Step 3: Making a hypothesis 2. Select the method and Steps in the design the study Step 4: Designing the study 3. Collect the data Research Process Step 5: Conducting the study 4. Analyze the data and draw conclusions Step 6: Analyzing the data 5. Report findings Step 7: Reporting the results 3 4 Step 1: Choosing a research question 5 6 Introduction Step 2: Conducting a literature review How to Find a Research Topic Step 3: Making a hypothesis Find inspiration from the world around you APA Step 4: Designing the study Informal Observations Method Style Step 5: Conducting the study Practical Problems Step 6: Analyzing the data Results & Previous Research Discussion Step 7: Reporting the results 5 6 1  9/16/2024 7 8 How to Find a Research Topic Scholarly Sources Do your own literature search to find scholarly research What makes an article a scholarly source? But how do you know what’s scholarly? Peer review Critical review of scientific merit and significance Two types of sources: Experts in the field 1. Articles in professional journals Suggestions for improving article 2. Scholarly books in psychology and related fields Recommendation for publication 7 8 9 10 Types of Scholarly Sources Where to Find Reliable Sources Journal Articles Journal articles typically School Library website Empirical papers include the following sections: Database search (PsycINFO) or directly from a Abstract journal Review articles Introduction Google Scholar Theoretical article Meta-analyses (effect size) Method Use the “Cited by” button to get more recent Results research that cites that article Chapter in an edited book Discussion Reference section of a relevant paper Full length book References Useful for older relevant research 9 10 Use the Library 11 Generating Research Questions 12 Questioning what you already know Make use of the library, while you have access! You just need to log in using your student # and password. Observation In psychology, you probably want to go to the library’s list of Public databases: Yourself https://guides.library.ontariotechu.ca/az.php?a=p Friends/family/pets PsycARTICLES (full articles) OR Nature PsycINFO (abstracts, often but not always linked to the Expanding upon previous research actual paper, from all psychology peer-reviewed papers) Improve someone else’s study Slides provided by the FSSH Librarian are available in Canvas as well Applying pre-existing theories 11 12 2  9/16/2024 13 14 Evaluating Research Questions Evaluating Research Questions Interestingness – Is the research question interesting to people in general, and the scientific community in Feasibility of answering the research question particular? Consider factors such as time, money, Three main factors influence the interestingness of a equipment/materials, technical knowledge and research question: skills, and access to participants 1. The answer is in doubt Do not waste time and effort pursuing research 2. The answer fills a gap in the research literature that you cannot complete successfully 3. The answer has important practical implications 13 14 What makes a good research 15 Theory vs. Hypothesis 16 question? Hypothesis Scientific method can be used to answer Theory Specific prediction about Parsimonious methods Cohesive an event or phenomenon Adds knowledge about the topic and isn’t completely explanation about a that is testable answered yet set of events that has Often but not always Have results which allow us to predict future outcomes not yet been shown derived from a theory to be untrue Inductive (many Has important practical outcomes observations  theory) or Passes the ROT test – Repeatable, observable, and deductive (theory  testable prediction) 15 16 Hypothetico-Deductive Method 17 What makes a good hypothesis? 18 Testable and falsifiable – Can be proven wrong Logical – Should be informed by previous theories/observations and logical reasoning Positive – Make a positive statement about the existence of a relationship Replicable – Can be repeated by someone other than the original experimenter and obtain the same results (e.g., Clever Hans) 17 18 3  9/16/2024 Step 1: Choosing a research question 19 20 Variables Step 2: Conducting a literature review Variable – A quantity or quality that varies across people Step 3: Making a hypothesis or situations Steps in the Step 4: Designing the study Quantitative Variable – A quantity that is typically measured by assigning a number to each individual. Research Process Step 5: Conducting the study Ex: Height, level of talkativeness, number of siblings Categorical Variable – A quality that is typically measure Step 6: Analyzing the data by assigning a category label to each individual. Ex: Occupation, whether someone is receiving Step 7: Reporting the results psychotherapy or not, university major 19 20 21 22 Operational Definitions Sampling and Measurement Operational Definition – A definition of the variable in terms Population: of precisely how it is to be measured. A large group of people about whom researchers in Most variables of interest cannot be directly observed or psychology are usually interested in drawing measured (e.g., depression). conclusions Operationally defining a variable takes it from an Sample: abstract concept to something that can be directly A smaller portion, or subset, of the population the observed or measured. researcher would like to study. Most variables can be operationally defined in many Should ideally be representative (i.e., similar in different ways. important aspects) to the population. 21 22 23 24 Sampling and Measurement Experimental Research Independent Variable (IV) – Variable manipulated by the Simple Random Sampling: experimenter (the presumed cause) Every member of the population has an equal Dependent Variable (DV) – Variable the experimenter chance of being selected for the sample. measures (the presumed effect) Convenience Sample: Extraneous Variables – Any variables other than the IV/DV The sample consists of individuals who happen to be Confounds – A specific type of extraneous variable that nearby and willing to participate (e.g., undergrad systematically varies with the variables under investigation students participating in research for course credit). Provide an alternative explanation for the results 23 24 4  9/16/2024 Non-Experimental Research 25 Laboratory vs. Field Research 26 Laboratory Study – Conducted in the laboratory The researcher simply measures variables as they environment and usually has high internal validity. naturally occur, but they do not manipulate them. Field Study – Conducted in the real-world (i.e., in the natural environment) and usually has high external validity. Non-experimental does NOT mean nonscientific. Internal Validity – The degree to which we can confidently infer a causal relationship between variables. Allows for description and prediction, but it does External Validity – The degree to which we can generalize NOT allow for making causal conclusions (i.e., the findings. cannot say that one variable causes the other with There is typically a trade-off between internal and external this approach). validity. 25 26 27 28 Analyzing the Data Measures of Central Tendency Descriptive Statistics: Mode – The most frequently occurring score in a Used to summarize the data distribution. Inferential Statistics: Mean – The arithmetic average of scores in a Used to generalize the results from the sample to the distribution obtained by adding the scores and then population of interest. dividing by the number of scores that were added Used to make conclusions about whether or not a theory together. has been supported, refuted, or requires modification. Median – The middle score in a rank-ordered We do not analyze the data in this course, but analyses do distribution. need to be considered when designing a study. 27 28 29 30 Example Measures of Dispersion Seven psychiatric in-patients had the following scores out of ten on a Describes the spread (or variation) of the scores in a distribution. depression inventory: EXAMPLE: You interview five people and ask them how old they 7 8 10 8 9 7 8 are What is the mode? What is the median? What is the mean? If they are all 28, there is no variation in the age of the people If they are ages 6, 20, 28, 37, and 49 then there is more First, arrange from lowest to highest value: 7, 7, 8, 8, 8, 9, 10 variation. Variability refers to how similar or different the people in the Mode = 8 sample are from one another. Two samples with the same mean Median = 8 can be very different. Mean = (7+7+8+8+8+9+10)/7 = 8.14 29 30 5  9/16/2024 Measures of Dispersion 31 Correlation Coefficient 32 Range – the difference between the highest and lowest When one trait or behavior accompanies scores in a distribution another, we say the two are correlated. Standard Deviation – a computed measure of how Indicates strength much scores vary around the mean of relationship (0.00 to 1.00) Correlation Coefficient r = + 0.37 The correlation coefficient (r) is a statistical measure of the Indicates direction relationship between two of relationship variables. (positive or negative) 31 32 Types of Correlation 33 Correlation is NOT Causation 34 (i.e., no correlation) 33 34 When are differences statistically 35 Type I and II Error 36 significant? H0 really is true H0 really is false When calculations indicate that research results are not likely to be result of chance, we say that the results are B statistically significant A Type II Error (β) You fail to reject H0 You correctly failed You failed to reject H0 Significance occurs when the probability of the observed to reject H0 when you should have findings being due to chance is very low rejected it Less than 5/100 (i.e., 5%) C Amount of variability is key Type I Error (α) D You rejected H0 Standard deviation is important You reject H0 You correctly rejected when you should H0 Depends on sample size: Larger sample size is better have failed to reject it 35 36 6  9/16/2024 37 38 Drawing Conclusions Given the possibility of Type I and II errors, we cannot use the results of a single study to conclude with certainty that a theory is true. Theories are supported, refuted, or modified based on the results of research. Because stats are probabilistic in nature and all studies have flaws, there is no such thing as scientific proof, there is only scientific evidence. 37 38 39 Assignment #1: Relevant Literature (15%) 40 Reporting the Results Due: Submit in Canvas by 11:59 pm on Friday, September 27th For this assignment, you will find 12 peer-reviewed articles that you believe are relevant to your research question. Include the abstract and The final step in the research process involves any necessary citation information for all 12 articles. reporting the results. For each article, write out a sentence or two in your own words (neither copying nor quoting allowed for this part!) explaining how this particular Results are typically reported in peer-reviewed article will be helpful to you when writing your literature review for journal articles and at conferences. Assignment 2. Presenting work at a conference is a great way to Include your research question at the top of Assignment 2, to give the TA get feedback from other researchers in the field the research context, and to keep you focused on your research question. before submitting it for publication. Instructions for Assignment 1, as well as a template, have been added to Canvas. 39 40 41 42 Creating Research Questions Research Proposal Another look at creating your research question For some context, here is a quick video that gives an https://www.youtube.com/watch?v=LWLYCYeCFak overview of how to write a research proposal (relevant You need to have a research question for Assignment #1 to Assignments 2 and 3): If you keep the same research question throughout https://www.youtube.com/watch?v=166FXhGd9T4 the course, you can use information from all your You do NOT need a timeline/schedule or budget smaller assignments when you write your research proposal The literature review (relevant to Assignment 2): You will have the opportunity to refine and improve https://www.youtube.com/watch?v=Vc_Yu_61Ymg the one question 41 42 7  9/16/2024 Questions about Chapter 2? 43 43 8