Midterm Flashcards by Mickey Scanlon

Four Goals of Scientific Research

To Describe Behavior
To Predict Behavior
To Determine Cause of Behavior
To Understand or Explain Behavior

How well did you know this?

Not at all

Perfectly

Producer vs Consumer

Producer - Take part in research process (studying, observing, conducting experiments, surveying)
Professor, lab volunteer, research scientist, etc.

Consumer - Read about research and apply to jobs or everyday lives
Therapists, counselors, teachers, etc.

How well did you know this?

Not at all

Perfectly

Sources of Research Ideas

Common assumptions
Observation of the world around us
Practical problems
Theories
Past research

How well did you know this?

Not at all

Perfectly

Source of information: Your Experience

No comparison group

Confounded - can not be sure what cause is

Only one point in overall pattern

How well did you know this?

Not at all

Perfectly

Source of information: Your intuition

Cognitive Biases - Swayed by a good story, Availability Heuristic, Present/Present Bias

Motivation Biases - Focus on evidence that supports our beliefs, Ask biased questions, Biased about being biased

How well did you know this?

Not at all

Perfectly

Research is Probabilistic

Scientific conclusions are based on patterns that emerge with multiple tests and comparison

Findings are not expected to explain all of the cases all of the time - only a certain portion of the possible cases

How well did you know this?

Not at all

Perfectly

The Intuitive Thinker vs the Scientific Reasoner

Intuitive thinking leads to mistakes

To counteract biases we need to adopt the empirical mindset of a researcher:
Base beliefs on systematic information
Strive to interpret data in an objective way

How well did you know this?

Not at all

Perfectly

Trusting Authorities on the Subject

Before taking advice of authorities ask yourself the source of their ideas:
Could be based on research but not all research is equally reliable
Could be based on experience
Could be based on intuition
Could be biased

How well did you know this?

Not at all

Perfectly

Anatomy of a Research Article: Abstract

Summary of entire research report/proposal

150-250 words

Includes: Hypothesis, procedure, broad pattern of results

How well did you know this?

Not at all

Perfectly

Anatomy of a Research Article: Introduction

Outlines the problems being investigated

Past research and theories relevant to the problems

Formal hypothesis or specific expectations are introduced and connected to past research

How well did you know this?

Not at all

Perfectly

Anatomy of a Research Article: Method

Subsections: Participants, materials, procedures, apparatus

Overview of design

Characteristics of participants

Procedure

Equipment and materials

How well did you know this?

Not at all

Perfectly

Anatomy of a Research Article: Results

How data will be examined

Findings presented in three ways:
Description in narrative form
Description in statistical language
Material in table or graphs

How well did you know this?

Not at all

Perfectly

Anatomy of a Research Article: Discussion

Review of research from various perspectives

Present methodological weaknesses or strengths

Explain how the results compare with past results

Includes suggestions for practical applications

Includes suggestions for future research on topic

How well did you know this?

Not at all

Perfectly

Basic vs Applied Research

Basic - attempts to answer fundamental questions about the nature of behavior with the goal of enhancing general body of knowledge

Applied - addresses issues in which there are practical problems and potential solutions, often guided by basic research

How well did you know this?

Not at all

Perfectly

Measured vs Manipulated Variables

Measured - levels are observed and recorded

Manipulated - levels are changed and controlled by researcher

How well did you know this?

Not at all

Perfectly

Independent vs Dependent Variables

Independent - levels are changed and controlled to test effect on dependent variable

Dependent - levels are tested and measured and suspected to be affected by the independent variable

How well did you know this?

Not at all

Perfectly

Operational Definition

Defining a conceptual variable at the theoretical level

Clear, concise, detailed definition of a measure

Operationalize

Turn a concept of interest into a measured or manipulated variable

Types of Measures

Self-report
Observational
Physiological

Empiricism

Conclusions are based on systematic observation

Characteristics of Scientific Approach

Make systematic observations and report them accurately

Engage in Theory-Building - development, testing, and refining of theories

Maintain an open system where falsifiable ideas are exchanged, debated, and challenged

Submit findings for peer-review

Theory

A set of statements that describes the general principles about how variables related to one another

Frequency Claim

Describe a particular rate or degree of a single variable

Focus on only ONE variable

Variable is measured (not manipulated)

Ex. 1 in 4 children enjoy potatoes

Association Claim

Argue that one level of a variable is associated with a particular level of another variable

Involves at least TWO measured variables (interval or ratio)

Correlated - as one variable changed the other tends to change also

Positive Association

High levels go with high levels, low levels go with low levels Ex. High score in badminton associated with high amount of practice

Negative Association

High levels go with low levels, and low levels go with high levels Ex. High scores on test associated with low amount of distractions during studying

Zero Association

No relationship among variables

Causal Claims

Argues that one variable is responsible for changing another variable

Criteria for Causal Claim

Covariation - must establish that one variable causes another Temporal Precedence - must establish the causal variable happened before the effect variable Internal Validity - Must establish that no other explanation exists for the relationship

Construct Validity

How well the variables in a study are measured or manipulated Extent to which the operational variables used in a study are a good approximation of the conceptual variables

External Validity

The extent to which the results of a study generalize to some larger population as well as to other times or situations

Statistical Validity

The strength of an effect and its statistical significance (the probability that the result could have been obtained by chance if there really is no effect) Extent to which a study minimizes the probabilities of two errors: Type I - False-Positive: Concluding that there is an effect when there is none Type II - Miss: Concluding that there is no effect when there is one

Internal Validity

The extent to which a confound is not responsible for the observed effect The relationship between one variable (A) and another (B) and the extent to which (A) rather than some other variable (C) is responsible for the effect on (B)

Validity

Appropriateness of a conclusion Valid claim - reasonable, accurate, and justifiable

Frequency Claim Validity

Construct Validity - How well was the variable measured Statistical Validity- How well do the numbers support the claim External Validity - Can we generalize from the sample to the population

Margin of Error

Statistical figure which indicates where the true value in the population probably lies

Association Claim Validity

Construct Validity of each variable - How well each were measured and constructed Statistical Validity - How strong is the association, is it statistical significant, were mistaken conclusions avoided External Validity - Can we generalize to the population as well as other contexts, times, or places

Statistical Significance

Likelihood that a relationship between variables is caused by something other than random chance

Statistical Significance: Mistaken Conclusion Errors

Error Type I - False-Positive - assuming a relationship when there is none Error Type II - False-Negative - (Miss) - assuming no relationship when there is one

Causal Claim Validity

Construct validity of both variables - how well were they measured or manipulated Statistical Validity - How big is the difference is it statistically significant Internal Validity - Are there other possible explanations External Validity - To who or what can we generalize this effect

Types of claims and variables

Frequency - 1 nominal variable Association - 2 interval or ratio variables Causal - 1 manipulated, 1 interval or ration variable