Introducing 

Prezi AI.

Your new presentation assistant.

Refine, enhance, and tailor your content, source relevant images, and edit visuals quicker than ever before.

Loading content…
Loading…
Transcript

Unit 1 Research Methods

Scientific Study

The Case for Scientific Study

The goals of psychological studies are to describe, explain, predict and perhaps influence mental processes or behaviors. In order to do this, psychologists utilize the scientific method to conduct psychological research.

The scientific method attempts to minimize the influence of bias or prejudice in the experimenter. Even the best-intentioned scientists can't escape bias. It results from personal beliefs, as well as cultural beliefs, which means any human filters information based on his or her own experience.

Cognitive Biases

Hindsight Bias: a psychological phenomenon in which one becomes convinced that one accurately predicted an event before it occurred. Also known as the "I-knew-it-all-along" phenomenon. Event has ocurred...and you look back.

Overconfidence: The Mother of all psychological biases...We think we know more than we do. We tend to be more confident than correct.

Predictable World Bias: Tendency to perceive order in random events. We have a built-in eagerness to make sense of our world

The Scientific Method

Scientific Method: a self-correcting process for evaluating ideas with observation and analysis

Scientific Theory: Explains behaviors by offering ideas that organize observations. Good theories explain by (a) organizing and linking observed facts, and (b) produces a testable hypothesis

Hypothesis: A testable prediction

  • If [Independent Variable], then [Dependent Variable]
  • Independent Variable is what the researcher manipulates
  • Dependent Variable is what the researcher measures

Research Methods

Experiment v. Correlational Study

Psychological studies vary in design. In correlational studies a researcher looks for associations or relationships among naturally occurring variables. Correlational studies include:

  • Survey
  • Naturalistic Observation
  • Case Study

Descriptive Research: Correlational studies DESCRIBE behavior only.

In experimental studies the researcher introduces a change (Independent variable) and then monitors its effects (dependent variable). An experiment is the only research method that can show a cause-and-effect relationship.

Survey

Survey: A descriptive technique for obtaining the self-reported attitudes or behaviors of a particular group

  • https://screening.mhanational.org/screening-tools/youth/?ref

Advantages: reach large amount of people, inexpensive, easy, multiple variables can be examined to estimate the attitudes or reported behaviors of a whole population

BUT...

  • Wording Effects: Asking questions is tricky, and the answers often depend on how questions are worded/framed
  • "Aid to the needy" v. "Welfare"
  • "Not allowing TV violence" v. "Censoring TV violence"
  • "Gun safety laws" v. "Gun control laws"

  • Social Desirability Bias: the tendency of survey respondents to answer questions in a manner that will be viewed favorably by others

  • Does not show causality

Naturalistic Observation

Observing and recording behavior in naturally occurring situations without trying to manipulate or control the situation

Advantage: "Real" behavior - humans don't live in a laboratory; offers interesting snapshots of everyday life

  • reduces...
  • Hawthorne Effect: the alteration of behavior by the subjects of a study due to their awareness of being observed

  • Demand Characteristics: participants form an interpretation of the experiment's purpose and subconsciously change their behavior to fit that interpretation

Disadvantage:

  • No control. May not be studying what you think you are studying - confounding variable?
  • Does not show causality

Case Studies

In-depth analysis of individuals or groups in the hopes of revealing some universal principles (things that apply to all of us)

Advantages: Can be revealing; can suggest direction for further study; opposite of a survey; good for rare disease/disorder; researchers can study something might be unethical to experiment

Disadvantages: Time consuming; Subjects are not representative; atypical participant (There is always an exception to the rule); Does not show causality

Longitudinal v. Cross Sectional Studies

Longitudinal Study: Research that follows and retests the same people over time

  • Advantages: uncover possible confounding variables; can be used in many research methods
  • Disadvantages: Time consuming, Expensive, difficult to control

Cross-Sectional Study: research that compares people of different ages at the same point in time

  • Advantages: comparison is science; can be used in other research methods
  • Disadvantages: only a snapshot, not a moving picture

Experimental Method

Experiment

Experiment: An investigator manipulates one or more factors (IV) to observe the effects on some behavior or mental process (DV). By Random Assignment of participants, the experimenter aims to control other relevant factors (confounding variables).

An experiment is the only research method that shows cause and effect.

Necessary elements of an experiment:

1. Manipulation of Independent Variable by researcher

2. Random Assignment of participants in to experimental and control groups.

Cause and Effect

  • Experiment is the ONLY research method that can demonstrate a cause and effect relationship between variables
  • Random Assignment is needed to demonstrate cause and effect.
  • The use of experimental controls reduces alternative explanations.

  • Correlational research can indicate if there is a realtionship of association between two variables but cannot demonstrate cause and effect.

Variables in an Experiment

Independent Variable (IV): Variable manipulated by the experimenter

  • Note: In a correlational study, IV is a designated predictor category (age/gender)

Dependent Variable (DV): Variable measured by the experimenter/researcher

Confounding Variable: anything (other than the manipulated IV) that affects change in the DV; unplanned. Confounding variables limit confidence in research conclusions.

  • Look to prompt (usually a clue). If in doubt:
  • memory, motivation, intelligence,
  • Placebo Effect
  • Experimenter Bias
  • Hawthorne Effect

Control Variable: any variable that's held constant in a research study. It's not a variable of interest in the study, but it's controlled because it could influence the outcomes. The use of controls reduces alternative explanations.

Identifying IV + DV Variable Practice

A high school principal is interested in proposing moving to a system in which some classes give participation points for speaking in class. She believes that for those classes in which participation points are given, students will enjoy those classes more. She is proposing an experimental study to examine this belief.

  • What is the Independent Variable? What is being manipulated?
  • What is the Dependent Variable? What is being measured?

Cora is a video game designer. She believes that games played using virtual reality headsets will be perceived by students as more challenging. She proposes an experimental study to examine this belief.

  • What is the Independent Variable? What is being manipulated?
  • What is the Dependent Variable? What is being measured?

Operational Definitions

Operational Definition: A carefully worded statement of the exact procedures used in a research study.

Operational Definitions are important because they allow for replication. For Dependent Variables, must have a measurable unit (seconds, feet, liters, %)

Ex: If I eat cookies before my exam, then my scores will improve

Ex: If I exercise, then I will become healthier

Many things we study in psychology are not directly observable. In order to measure these internal characteristics, needs to be operationalized (so we know what was manipulated and measured).

Choosing Participants

Population: All those with specified characteristics

Target Population: Subgroup of population you want to study and will generalize results to

Sample: the group of people who take part in the experiment/research investigation

Random Sampling

Random Sample: Each member of the target population had an equal chance of being chosen for the experiment/investigation

Random Selection: How you practically ensure a random sample. Refers to how participants are selected from the target population for inclusion.

  • Gold Standard = randomized computer generator

Representative Sample: We want out sample to match the "demographics" of the target population. Using random selection does not guarantee a representative sample.

  • Sample Size is important, small sample size can lead to a sampling bias and inhibit generalization

Generalizability and Sampling Bias

Generalizability: degree to which results of a study based on a sample can be said to represent the results that would be obtained from the target population from which the sample was drawn.

Sampling Bias: errors in research studies when the researchers do not properly select their participants

  • Non-Random Sample: volunteers are a red flag
  • Unrepresentative Sample: do the "demographics" of the sample match those of the target population?
  • Sample Size: small sample sizes are a red flag
  • False Consensus Effect: We tend to overestimate how many people share our judgments, values, choices...birds of a feather flock together

Experimental Design

Independent Measures/Between-Subjects Design: Each condition of the experiment includes a different group of participants

Repeated Measures/Within-Subjects Design: The same participants take part in each condition of the experiment; Control First!

Matched Pairs: Each condition uses different but similar participants

Experimental v. Control Groups

Experimental Group: the group of participants exposed to one version of the independent variable

Control Group: the group of participants not exposed to the independent variable; may receive placebo

Control Condition: The control condition in an experimental design lacks any treatment or manipulation of the independent variable. Everything in a control condition is the same as the experimental conditions except that the independent variable is absent or held constant.

Participants are placed in either Experimental or Control Group through Random Assignment. Random Assignment is necessary to show cause-and-effect.

Random Assignment

How are participants placed into the Experimental Group or the Control Group?

Random Assignment: assigning participants to experimental and control groups by chance, thus minimizing the pre-existing differences between different groups

  • Needed to demonstrate cause and effect
  • reduces the effect of confounding variables

Random Assignment v. Random Selection:

Random selection refers to how the sample is

drawn from the population as a whole, while

random assignment refers to how the

participants are then assigned to either the

experimental or control groups. It is possible

to have both random selection and random assignment

in an experiment

Does Coffee Cause Stellar Test Score?

You score your highest grade ever on the AP Psych test. You think back to what you may have done to achieve this good grade and you remember that you drank a cup of caffeinated coffee on the morning of the test - you definitely feel this made the difference. You decide to conduct an experiment. You select students for the study and assign 1/2 to the control group and 1/2 to the experimental group.

  • What is the IV? Operational Definition?
  • What is the DV? Operational Definition?
  • What is the Experimental Group?
  • What is the Control Group?
  • Any confounding variables?

Demand Characteristics and Experimenter Bias

Demand Characteristics: the clues that participants discover about the purpose of the study, including rumors they hear about the study suggesting how they should respond

  • Fix --> Single-Blind Procedure: researchers know which participants have been placed in the control and which participants have been placed in the experimental group, but the participants are uninformed about the treatment (if any) they are receiving

Experimenter Bias: phenomenon that occurs when a researcher's expectations or preferences about the outcome of a study influence the results obtained

  • Fix --> Double-Blind Procedure: Both the research participants and the researchers are blind to which participants are in the experimental group and which participants are in the control group

Placebo Effect

Experimental results caused by expectations alone

Sometimes just THINKING you are getting treatment can boost your spirits, relax your body and relieve your symptoms

Research Methods Review

Test Yourself

Statistical Analysis

Statistics

Statistics: tools that help the psychologist to interpret the vast quantities of information they are confronted with on a daily basis

Descriptive Statistics: numerical data used to measure and describe the characteristics of groups. Describes data. Tells us nothing about statistical significance.

  • Measures of Central Tendency - Mean, Median and Mode
  • Measures of Variation - Range and Standard Deviation

Inferential Statistics: data that allows you to make predictions (“inferences”) from that data. With inferential statistics, you take data from samples and make generalizations about a population. A "p-value" will indicate whether results are statistically significant.

  • When P is less than .05 results are statistically significant (ok with this much chance)
  • When P is more than .05 results are not statistically significant (not ok with this much chance)

Statistical Significance: results are due to the manipulation of the IV and NOT due to chance; indicates likelihood that a result could have happened by chance

Descriptive: Measures of Central Tendency

12, 3, 3, 4, 8

  • Mean?
  • Mode?
  • Median?

The Median and the Mode are relatively unaffected by outliers (extreme scores). One extreme score will throw off the Mean.

Descriptive: Measures of Variance

Range: the difference between the highest and lowest scores in a distribution

Standard Deviation: a measure of how much scores vary around the mean score. Just means how far out from "average". A large standard deviation indicates that the data points are far from the mean, and a small standard deviation indicates that they are clustered closely around the mean.

Relationship between Range and Standard Deviation

  • The larger the range, the larger the standard deviation

Interpreting Graph: A test has a mean of 52 with a standard

deviation of 12. What scores are within one standard deviation

of the mean?

  • Mean always in the middle
  • Hash marks to left and right are the standard deviations from the mean.

Calculating Standard Deviation

Step 1: Find the Mean

Step 2: Determine the deviation from the Mean (how far away that number is from

the mean) for each

Step 3: Square the deviations

Step 4: Sum of deviations squared (add them up!)

Step 5: Divide the sum of the deviations squared by the number of scores (the

Variance)

Step 6: Take the square root of Variance

Relationship between Standard Deviation and Variance. Variance measures how far each number in the set is from the mean and thus from every other number in the set.

  • Standard Deviation squared = Variance
  • Square root of Variance = Standard Deviation

Normal Distribution

A normal distribution comes with a perfectly symmetrical shape. This means that the distribution curve can be divided in the middle to produce two equal halves. The symmetric shape occurs when one-half of the observations fall on each side of the curve.

  • A symmetrical, bell-shape curve
  • Mean = Median = Mode

Bimodal Distribution

A bimodal distribution is a set of scores with two peaks or modes around which values tend to cluster, such that the frequencies at first increase and then decrease around each peak.

Skewed Distributions

Positively Skewed Distribution: Long tail in the positive direction; Mean is greater than the Mode and Median

  • Test too hard

Negatively Skewed Distribution: Long tail in the negative direction; Mode and Median are greater than the Mean

  • Test too easy

Correlational Methods Review

Correlation: A measure of the extent to which two factors vary together, and thus of how well either factor predicts the other.

Predicted Relationship: if there is a relationship between two variables, we can make a prediction as to the direction of that relationship. Similar to the hypothesis of an experiment

Correlational Studies: survey, naturalistic observation and case study

Measuring and Depicting Correlations

Correlation Coefficient: a statistical index of the relationship between two things (from -1.00 to 1.00);

  • shows STRENGTH and DIRECTION of relationship
  • +/- indicate direction (positive or negative)
  • number indicates strength; closer to 1 stronger, closer to 0 weaker

Scatterplots are used to plot data points on a horizontal and a vertical axis in the attempt to show how much one variable is affected by another.

X axis = "Independent Variable" - designated category

Y axis = Dependent Variable

Positive, Negative and No Correlation

No relationship or connection between the two variables

Score with a negative (-) sign; means there is an inverse relationship - as one behavior increases, the other decreases

Score between +0.01 and +1. The two traits/behaviors either increase or decrease together; direct relationship

Correlational Cautions

Third Variable Problem: the fact that an observed correlation between two variables may be due to the common correlation between each of the variables and a third variable rather than any underlying relationship (in a causal sense) of the two variables with each other.

Illusory Correlation: The perception of a relationship where none actually exists. We can easily deceive ourselves by seeing what is not there.

Regression Toward the Mean: the tendency for extreme or unusual scores or events to fall back (regress) toward the average. Average results are more typical than extreme results.

When two variables, a and b, are found to be positively or negatively correlated, it does not necessarily mean that one causes the other: It may be that changes in an unmeasured or unintended third variable, c, are causing a random and coincidental relationship between the two variables by independently changing a and b.

CorrelationsTest Yourself

Ethics

Ethics in Psychological Research

Ethics refers to the correct rules of conduct necessary when carrying out research. We have a moral responsibility to protect research participants from harm. The purpose of these codes of conduct is to protect research participants, the reputation of psychology, and psychologists themselves.

Institutional Review Board: This group review serves an important role in the protection of the rights and welfare of human research subjects. The purpose of IRB review is to assure, both in advance and by periodic review, that appropriate steps are taken to protect the rights and welfare of humans participating as subjects in the research

American Psychological Association Ethical Rules

Informed Consent (before): the process by which researchers working with human participants describe their research project and obtain the subjects' consent to participate in the research based on the subjects' understanding of the project's methods and goals

Debriefing (after): is an essential part of the consent process and is mandatory when the research study involves deception. The debriefing provides participants with a full explanation of the hypothesis being tested, procedures to deceive participants and the reason(s) why it was necessary to deceive them.

Confidentiality: refers to the researcher's agreement to handle, store, and share research data to ensure that information obtained from and about research participants is not improperly divulged

Absolute Right to Withdraw: a study participant in a clinical trial has a right to end participation in that trial at will.

APA Ethical Rules Continued

No undue stress or harm: "Undue stress" refers to any stress that would be more than a participant would encounter in everyday life. Simple discomfort or embarrassment is not "undue stress or harm."

No Deception (if deception, must debrief): refers to misleading or tricking participants about the purpose or direction of the study.

APA sets the rules...the IRB ensures the rules are followed and approves research.

Federal Laws - Focus on Human Protection

Federally Funded Research may have additional requirements imposed by federal law.

  • The Federal Policy for the protection of human subjects requires that each institution "engaged" in Federally-supported human subject research file an Assurance with the Office of Human Research Protections. The Assurance formalizes the institution's commitment to protect human subjects.

  • IRBs reviewing clinical investigations involving Food and Drug Administration (FDA) regulated products are required to register with the FDA.

  • US Dept. Health Human Services - institutions engaging in most HHS supported human subject research must comply with protective rules

  • HIPAA - creation of national standards to protect sensitive patient health information from being disclosed without the patient's consent or knowledge

  • Protections for Children participants - like parental consent

Animal Research

  • Clear Scientific Purpose
  • Humane Care + Enriching Environment
  • Minimize Discomfort
  • Test painful stimuli on psychologist first

Emory University - IACUC - http://www.iacuc.emory.edu/ (under federal mandate to monitor all research activities related to animal care)

Learn more about creating dynamic, engaging presentations with Prezi