Chapter 2 Flashcards by Keira Niles

What are the four goals of scientific research in psychology?

To describe, predict, explain and control behavior

How well did you know this?

Not at all

Perfectly

How do psychologists describe behavior?

This is the first step in identifying behavior in the workplace (ex: describing how employees spend their time)

There are more and less observable behaviors that can be described, meaning that some may go unnoticed.

How well did you know this?

Not at all

Perfectly

How do psychologists predict behavior?

Through understanding how a target behavior is related to other factors

How well did you know this?

Not at all

Perfectly

How do psychologists explain behavior?

In order to change behavior, we must know what causes it, evidence to aid in this goes beyond description/prediction.

How well did you know this?

Not at all

Perfectly

Three types of evidence used ot explain cause and effect of behavior:

Temporal precedence
Covariation of cause and effect
Alternative explanations must be eliminated

How well did you know this?

Not at all

Perfectly

Temporal precedence

cause that precedes the effect

How well did you know this?

Not at all

Perfectly

Covariation of cause and effect

when the cause is present, so is the effect. When the cause is not present, neither is the effect

How well did you know this?

Not at all

Perfectly

Alternative explanations must be eliminated to explain behavior because…

even if other factors are involved, they are not what is driving the behavior if the cause and effect are truly correlated.

How well did you know this?

Not at all

Perfectly

What does it mean to say that psychologists control behavior?

How to control or manipulate phenomena, typically there are any factors that contribute to behavior

How well did you know this?

Not at all

Perfectly

Research Question

The guiding question behind a research study

How well did you know this?

Not at all

Perfectly

Hypothesis

researchers best guess at results of a study

How well did you know this?

Not at all

Perfectly

Theory

set of interrelated concepts and propositions that provide a systematic view of a phenomenon (based in research and evidence)

How well did you know this?

Not at all

Perfectly

What makes a good theory? (5)

Parsimonious (simple)
Precise (specific and accurate)
Testable
Useful
Generative (stimulates future research)

How well did you know this?

Not at all

Perfectly

Inductive reasoning

data –> theory

How well did you know this?

Not at all

Perfectly

Deductive reasoning

developing ideas from theory –> to test with data

How well did you know this?

Not at all

Perfectly

Stages of the research process (and where inductive/deductive reasoning would occur)

Formulate hypothesis (deductive) > design study > collect data > analyze data (inductive) > report findings >

How well did you know this?

Not at all

Perfectly

Observational method (+ pros and cons)

relationships are studied by observing variables of interest (correlational method)

Most useful for describing/predicting behavior.

Cannot determine direction of cause and effect, there may also be 3rd/confounding variables that influence cause and effect that the researcher cannot identify using this method.

How well did you know this?

Not at all

Perfectly

Types of observational methods: (4)

naturalistic observation
case studies
archival research
surveys

How well did you know this?

Not at all

Perfectly

Naturalist observation

Observational method by observing someone/something in its natural environment

How well did you know this?

Not at all

Perfectly

Case studies

Observational method by examination of a single person, group, company, society, etc.

How well did you know this?

Not at all

Perfectly

Archival research

Observational method by use of existing/secondary datasets

How well did you know this?

Not at all

Perfectly

Surveys

Observational method whereby surveys are self administered, based on interviews/drawn from experience sampling methods

How well did you know this?

Not at all

Perfectly

Experimental methods

where the researcher manipulates the independent variable and controls the other variables

How well did you know this?

Not at all

Perfectly

Independent variable

manipulated

How well did you know this?

Not at all

Perfectly

Dependent variable

the outcome

Internal validity

ability to draw conclusions about causal relationships within a study

Internal validity is related to...

experimental control and random assignment

Experimental control

every feature is held constant, except the independent variable

Random assignment

all other variables are controlled by random assigning participants into experimental/control groups

External validity

extent to which studs results can be generalized to other populations/settings.

Artificial lab experiments may have...

limited external validity

What two experiments allow for increased external validity?

Field and quasi experiments

Field experiments

occur in natural settings with some degree of control

Quasi experiments

also conducted in real world settings but lack random assignment of participants to groups

Importance of replication is science

replication = repeating a study should produce the same results consistently

3 types of activity that jeopardize science

1. Research misconduct (plagiarism/falsification) 2. Bad research practices (p-hacking: making statistically significant correlations seem like they are linked when they actually hold no meaning 3. Other misconduct (misreporting findings in the media)

Psychological measurements for surveys (2)

constructs and items

constructs

most are unobservable (things like depression, conscientiousness) so we use items to measure constructs

Items

what we use to measure a construct (a series of questions)

Reliability

the consistency/stability of a measure

true score

real value of a given variable

Measurement error

difference between the measured score and the true score

Types of reliability (4)

test-retest parallel forms internal consistency interrater

Test-retest

measuring same construct at two points and comparing results

Parallel forms

uses two different forms of the test, instead of repeating the same test. Avoids issue of participants remembering earlier responses. SAT/ACT is an example of this

Internal consistency

reflects how well items on a test hang together. Because all items measure the same construct, they should be highly correlated

Interrater

correlation between observations of different raters. Increased correlation means raters agree in their ratings.

Reliability vs validity

reliability refers to predictability/consistency of outcomes while validity refers to accuracy (hitting a bullseye)

Construct validity

when something measures what it is intended to measure. There are 6 types

What are the 6 types of construct validity?

face content predictive concurrent convergent discriminatory

face validity

content of a measure appears to reflect the construct (very superficial) Questions cannot be too obvious to where people know what you are measuring

content validity

content of the measure fully covers the construct domain

predictive validity

scores predict a criterion measured at a future time (future job performance)

concurrent validity

scores are related to a criterion measured at the same time (current job performance)

convergent validity

scores similar to other measures of the same construct

discriminant validity

scores are not related to measures of theoretically different constructs

Descriptive statistics

statistical measures that allow precise statements about data (statistics can be misleading)

Measures of central tendency

values that describe the typical/central score among a set

variability

amount of spread in a distribution of scores

Correlation coefficient

symbolized as "r" reflects how strongly two variables are related to each other (-1.0 to 1.0)

Regression

statistical technique that allows us to predict scores on one variable based on the scores from another variable (step further than correlations)

Meta analysis

a study design and statistical technique that combines data from many studies

how does meta analysis work?

compile all studies on a given topic, then analyze results alongside each other. Gives an estimate of the average effect and how it differs from moderator variables

Advantages of meta analysis

- Can avg effects across an entire literature - can ask questions which were not the primary focus

Disadvantages of meta analysis

- publications bias (file drawer problem - scientists are more likely to post positive results over negative ones) - garbage in, garbage out

Daisy believes that by studying the reasons why her cats scratch, she can ultimately rearrange her furniture to change her cats’ scratching behaviors. This ultimate goal fits with which of the following goal of science? a. Observation b. Prediction c. Exploration d. Control

Daisy observed her five cats for 10 days to collect data and then tried to develop a theory of her cats’ scratching behavior. She is following the ______ approach. a. Production b. Reduction c. Inductive d. Deductive

After the initial data collection, Daisy came up with a theory that when cats are stressed, they are more likely to scratch the furniture. In this case, stress in cats is the ______ variable. a. Dependent b. Extraneous c. Control d. Independent

Kevin gave a test to a group of students at time 1, and then gave the same test three months later. He found that the high (and low) scorers on the first test were also the high (and low) scorers three months later. This test has good: a. Content validity b. Test-retest reliability c. Criterion-related validity d. Internal consistency

Job applicants’ personality test scores were strongly related to their job performance. This means that the test has good: a. Content validity b. Test-retest reliability c. Criterion-related validity d. Internal consistency

For the exam 1 for I/O psychology class, Kevin accidentally included a question from a chapter that is not yet covered in the class. This means that: a. The test has high construct validity. b. The test has low content validity. c. The test has high predictive validity. d. The test has low concurrent validity.

Other than the one bad item, Kevin found that all the other items for exam 1 for this class are highly related and hang together. This means that: a. The test has high discriminant reliability. b. The test has low content validity. c. The test has high internal consistency. d. The test has is low divergent validity.

Daisy decided to test her theory about cat stress and furniture scratching using a quasi-experimental method with 20 cats. This means that ______. a. she will randomly assign the 20 cats into two groups to be exposed to stressed versus calm conditions. b. she will observe the 20 cats in their natural condition. c. she will use existing grouping of cats (i.e., 5 at her place, 15 at a near-by shelter) and expose them to calm versus stressed conditions