Lecture 2: RESEARCH METHODS PDF

Preview

Summary

Lecture 2: RESEARCH METHODS Lecture 2: RESEARCH METHODS...

Description

9/15/21

Audio Recordings of Lectures • Audio recordings of lectures now posted at:

lectures.its.sfu.ca Scroll down the page to find “PSYC 102-100”

Study Tip of the Week • Research shows that spaced studying is more effective than

massed studying for learning and retention • This week, do two things: 1. 2.

Schedule 2-3 different study sessions exclusively for this class Time how long it takes you to do one task (e.g., figure out how long it takes to read the chapter. Use this info to make your study plan for next week)

4 hours of spaced studying

4 hours of massed studying (“cramming”)

1

9/15/21

The Research Enterprise in Psychology Chapter 2 Dr. Diana Lim Email: [email protected]

3

Learning Objectives • List the steps in scientific method in the correct order • Give an example of the scientific method in action • Describe the advantages of the scientific method

2

9/15/21

4

Really? • Why does psychology need the scientific method? Isn’t

common sense good enough? • Problem: Common sense is often contradictory…

Opposites Attract! Birds of a feather flock together

5

Really? • Why does psychology need the scientific method? Isn’t

common sense good enough? • Problem: We are biased by our own experiences, and tend to

see what we expect to see…

3

9/15/21

6

Really? • Problem: There’s a LOT of information out there…How do we

know what to believe?

7

So… • What distinguishes science from common sense? • What is the difference between psychology and “pop psychology”?

4

9/15/21

8

Steps in the Scientific Method Evaluate and/or adjust your hypothesis and continue!

In a journal, at a conference, etc.

1. Formulate a (testable) hypothesis

5. Report the findings

2. Design the study

4. Analyze the data

Consider ethics; Be careful of sampling bias

3. Collect the data

Be careful of placebo effects, experimenter bias, or distortions in self-report

9

Steps in the Scientific Method I’ll consider my results and ask my next question

I’ll share my findings with my lab. If they are strong enough, I may try to submit to a journal for publication

1. Formulate a (testable) hypothesis

5. Report the findings

I’ll use stats to compare groups and see if there is a meaningful difference in performance

4. Analyze the data

Hyp: Drinking a cup of coffee will improve performance on a memory task

2. Design the study

3. Collect the data

I’ll use two groups: one group will get coffee and the other will get water

I’ll have all participants do the memory task at the same time, and then I will score their accuracy

5

9/15/21

10

Steps in the Scientific Method Viewing sexualized pictures lead to greater body dissatisfaction and predicts greater desire for cosmetic surgery (published in journal “Body Image”)

1. Formulate a (testable) hypothesis

5. Report the findings

Compare body dissatisfaction scores before and after viewing images

Hyp: Viewing images of sexualized women on social media leads young women to feel dissatisfied about their bodies and desire cosmetic surgery

2. Design the study

4. Analyze the data

3. Collect the data

Young women given a survey on body dissatisfaction, then view images of “perfect” bodies, then take the survey again

All women are given the same survey, and view the same images. A control group will view nonsexualized images

Full article: Guizzo, Canale & Fasoli (2021). Instagram Sexualization: When posts make you feel dissatisfied and wanting to change your body. Body Image, 39(62-67)

11

The Scientific Method • The scientific method includes: • Being precise about what is being measured and how it will be measured (“operational definitions”) • Using statistics to evaluate data in an objective way

How do you measure height?

How do you measure “intelligence?”

How do you evaluate the data to come to conclusions?

6

9/15/21

12

The Scientific Method • What happens if you aren’t precise with defining your variables?

Video: https://www.youtube.com/watch?v=FN2RM-CHkuI

13

The Scientific Method • Advantages of using the scientific method

It is inherently precise 2. It is intolerant of error 1.

7

9/15/21

14

Practice • Create operational definitions for the following research

questions: • Does marital satisfaction increase with age? • Does warm weather correlate with drug use? • Does group therapy help people with social anxiety?

• Remember: An operational definition is the statement of procedures the

researcher is going to use in order to measure a specific variable.You will need an operational definition for each of the variables above (e.g., for the first statement, you need to operationally define “marital satisfaction” and “age”)

Research Methods What kinds of approaches can we use to answer psychology questions?

8

9/15/21

16

Learning Objectives • Explain three research methods

used to find or describe links 1. 2. 3.

Naturalistic observation Case studies Surveys

• Explain how a true experiment

examines cause and effect • Identify and define important terms

used in experiments • Independent /dependent variables • Control • Extraneous variables

17

Naturalistic Observation • Researcher carefully and systematically observes and records

behaviour without interfering with the behaviour • Natural (“in the field”) or in the laboratory

Chimpanzee ‘fishing’ for termites Jane Goodall, circa 1963

9

9/15/21

18

Case Study • A detailed description of a particular individual being studied or

treated based on careful observation or psychological study • Also known as case history

Phineas Gage

19

Surveys • Questionnaires and interviews that gather information by

asking people about their opinions, experiences or attitudes

10

9/15/21

20

Review: Descriptive Studies • Looking for links • Cannot test cause and effect with these methods Naturalistic observations

Case studies

Surveys and interviews

• Records behavior in natural environment (or in the lab)

• Examines one individual (or case) in depth

• Describes (but does not explain) behavior

• Provides fruitful ideas

• Examines many cases in less depth • Careful of wording effect • Low rate of return

• Can be revealing (but difficult to control)

• Cannot be used to generalize

• Use random sampling of population for best results

Myers et al. (2016). Psychology, 11th Edition

21

What is an Experiment? Experiment: manipulation of one variable under controlled conditions so that resulting changes in another variable can be observed

•

•

Detection of cause-and-effect relationships

Variable 1 = drink

Variable 2 = performance on memory task

11

9/15/21

22

Independent and Dependent Variables Independent variable (IV) = variable manipulated • Dependent variable (DV) = variable affected by manipulation • Detection of cause-and-effect relationships •

23

Extraneous Variables • Any variable other than the IV that seem likely to influence the

DV in a study

12

9/15/21

24

Handling Extraneous Variables • Add control • Create a more complex design (e.g. with 2 independent

variables)

25

Practice Let’s say you are interested in how artificial red light affects memory. • What is your hypothesis? (Be specific and create operational

definitions for all variables!) • Design an experiment to test this hypothesis • What are the independent and dependent variables? • What will happen to the experimental group? • What will happen to the control group? • What extraneous variables may be present? • How will you control for extraneous variables?

13

9/15/21

Statistical Tools for Psychology Research How do we know if our results are meaningful and not just a coincidence or a fluke?

27

Learning Objectives • Give a brief argument for why psychology needs statistical

tools • Distinguish between descriptive statistics and inferential

statistics • Determine which type of statistic is appropriate for different scenarios

• Explain the concept of “significance” in psychological studies

14

9/15/21

28

Statistics in Experimental Design • Why do we need statistics?

More bad graphs: https://www.distractify.com/humor/2017/05/26/cszBB/hilariously-bad-graphs

29

Statistics in Experimental Design • Why do we need statistics? • Statistics help us organize, summarize, and interpret numerical data • What do you need to know? • Descriptive statistics are used for summarizing data (e.g. mean, median,

mode) • Inferential statistics are used for interpreting data and drawing conclusions (e.g. hypothesis testing)

15

9/15/21

30

Describing Data • Psychologists often create graphs to represent their results.

The most common kind is the frequency distribution • A normal distribution is called a bell-curve

Myers et al. (2016). Psychology, 11th Edition

31

Descriptive Statistics • Descriptive statistics are used to summarize the data • Measures of central tendency • Mean: the average value (mathematically) • Median: the value in the middle of the range • Mode: the value of the most common score

From Schacter et al. (2014), Psychology 3rd Edition

16

9/15/21

32

Descriptive Statistics • Measures of variability: • Range: the largest measurement in a distribution minus the smallest measurement • Standard deviation: the average difference between the measurements in a frequency distribution and the mean of the distribution. • Percentile score: indicates the % of people who score at or below the score

33

Descriptive Statistics • Correlation: A correlation exists when two variables are

related to each other • Correlation coefficient: A numerical index of degree of relationship; when two variables are related to each other • Correlation expressed as a number between 0 and 1 • Can be positive or negative • Numbers closer to 1 (+ or –) indicate stronger relationship

17

9/15/21

34

Correlation: Slope Direction • The direction (slope) of the correlation indicates the nature of

the changes in the variables

Negative correlation

No correlation

Positive correlation

35

Correlation: Strength • The degree of association between the two variables is the

strength of the correlation, expressed as a correlation coefficient (r)

18

9/15/21

36

Correlations Do Not Refer to Individuals

37

Correlation ≠ Causation • When two variables are related to each other, they are

correlated • But… Correlation does not mean causation

19

9/15/21

38

Correlation ≠ Causation • When two variables are related to each other, they are

correlated • But… Correlation does not mean causation

Directionality problem

A

B

A

B

A Third variable problem

C B

39

Inferential Statistics • Inferential statistics are used to: • Interpret data • Draw conclusions • Make predictions

• For example… • Let’s say, on average, the caffeine group remembers 20 items and the

water group remembers 17 items. What does this mean?? • You need more context in order to draw a conclusion • You can use this to make future predictions • E.g., If I did this again, I think I’d get similar results

20

9/15/21

40

Statistical Significance • In psychological research, a result is considered significant if

the odds of it occurring by chance are very low (typically, less than 5%) • Usually, greater than 2 standard deviations away from the mean • Reported as 0.05 level of significance (p < 0.05)

41

Statistical Significance • Be skeptical • Stats test don’t really give us a definitive answer - there is always a chance that the results were a fluke • It is possible to use statistics in a biased way • Statistical significance is not the same as practical (or clinical) significance

21

9/15/21

42

Study #1:Video Games A team of psychologists want to determine if playing violent video games causes increased aggression in children. They gather a random sample of 1,000 children in BC, aged 8-12. Each child completes a survey to determine their aggressive tendencies. Each child also reports the number of hours playing video games. What are the variables in this study? 2. What are the advantages and disadvantages of using the survey method? 1.

3. Will this study show correlation or causation? Hint: ask yourself whether anything is being manipulated. Manipulation of a variable (by the researcher) is a sign of experimentation, not correlation.

43

Study #2:Talking About Stereotypes Researchers want to investigate the stereotype that women talk more than men. They give 20 college students recording devices to wear that will automatically record them at random intervals throughout the day. The recorders collect a sample from each individual for the researchers to compare. What are the variables? 2. How could this experiment be improved to reduce bias? 1.

22

9/15/21

44

Study #3: Emotions and attraction A group of researchers are interested in how fear and stress relate to attraction in young men. They design an experiment where participants are either put in a stressful situation (crossing a high suspension bridge) or a non-stressful situation (crossing a low bridge). After crossing the bridge, participants are greeted by a young woman and asked about their experience. After this experience, participants are asked to rate their attraction levels via questionnaire. What are the variables in this experiment? 2. Are there any confounding variables that may affect the results? 3. Will this study show correlation or causation? 1.

Hint: ask yourself whether anything is being manipulated. Manipulation of a variable (by the researcher) is a sign of experimentation, not correlation.

Research Ethics What can and can’t be done in psychological research?

23

9/15/21

46

Poll Break

47

Learning Outcomes • Define research ethics and provide

an argument for why they are necessary in psychology research • Outline the ethical guidelines for

psychology research • In human research • In animal research

24

9/15/21

48

Research Ethics • Ethics: the systematic study of values, so as to decide what is

the proper action and/or conduct • Two major responsibilities in research ethics 1.

Ensure the welfare, dignity, and safety of both human participants and non-human research subjects

2.

Ensure that public reports are honest and accurate

49

The Challenge • The goal of clinical and psychological research is the generation

of useful knowledge about human health and behaviour • Problem: People are the means to developing useful knowledge, and thus

are at risk of exploitation • How do we avoid “the ends justify the means” mentality?

25

9/15/21

50

Ethical Guidelines • Canadian Psychological Association

51

Ethical Guidelines • In psychology research: • Participants must be protected from harm • Participants must be informed of the benefits and the risks • If deception is required, participants must be debriefed afterwards • Participants must give informed consent • Participant information must be kept confidential

26

9/15/21

52

Why do Animal Research? • Gain knowledge about animals • Discover practical applications • Explore questions that can’t be studied with humans for

practical or ethical reasons • Improve animal and/or human welfare

53

Ethical Guidelines:Animal Research • The Canadian Council on Animal Care (CCAC) sets

standards for animal ethics and care in science within Canada • Every experiment with animals must be approved by the Canadian Council on Animal Care (CCAC) • The CCAC also provides training for researchers who will be handling or working with animals

27

9/15/21

54

Ethical Guidelines:Animal Research • The three “R’s” of animal research 1. Replace animal models with methods 2. Reduce the number of animals used 3. Refine experimental procedures to improve animal welfare

55

Next Steps • Required: • Read Chapter 2 of your textbook (if you haven’t yet) • Optional • Post any additional questions on the Discussion Board on Canvas • Attend virtual office hours to ask questions or review your Ch 1 quiz

• When you’re ready…. • Complete the Chapter 2 quiz before next class • Next class: Chapter 8

28

9/15/21

56

Summary: Comparing Research Methods • Instructions: Fill in the blanks of the table Research Method

Basic Purpose

Experimental

What Is Manipulated (if anything)

Do case studies, naturalistic observations, or surveys

Descriptive

Correlational

How Conducted

To detect naturally occurring relationships; to assess how well one variable predicts another

Weaknesses

No control of variables; single cases may be misleading

Nothing

Manipulate one or more factors; use random assignment

Sometimes not feasible; results may not generalize to other contexts; not ethical to manipulate certain variables

Myers et al. (2016). Psychology, 11th Edition

57

29...