Chapter 5 Correlational And Quasi Experimental Design

Front 1

2 Categories of Non experimental Research Methods

Back 1

Correlational designs

Quasi-experimental designs

Front 2

Establish relationships among pre-existing behaviours and predict one set of behaviours from others.

Eg, college grades prediction from entrance exam score

Back 2

Correlational designs

Front 3

Quasi "Latin meaning"

Back 3

Seeming like

Front 4

Lack of manipulation of antecedents or random assignment and lack of treatment conditions

Back 4

Quasi-experiment

Front 5

inability to establish cause with certainty in research.

Back 5

Confounding

Front 6

Low of Manipulation of Antecedents

Back 6

----

Front 7

Vary in degree of manipulation of antecedents, but random assignment

Back 7

Quasi -experiments

Front 8

Restrict, or limit the responses subject can contribute to the collected data.

Back 8

Imposition of units

Front 9

Design to determine the correlation, or degree of relationships between two traits, behaviours or events

Back 9

Correlational study

Front 10

Refers to any observable behavior, or characteristics or event that can vary or have different values

Back 10

Variable

Front 11

Correlational designs

Back 11

---

Front 12

Design to determine correlation, or degree of relationships between two traits, behaviours and events.

Back 12

Correlational study

Front 13

Relationship between pairs of scores from each subject

Back 13

Simple correlations

Front 14

Most commonly used procedure for calculating simple correlations can result in three (3) general outcomes: a positive relationship, negative relationship, no relationship

Back 14

Pearson product moment correlation coefficient (R)

Front 15

Visual representation of the scores belonging to each subject in the study.

Back 15

Scatterplots

Front 16

(Lines of best fit)

Describes the linear relationship between the two measured scores and illustrate mathematical equation.

Lines drawn on scatterplots.

Back 16

Regression lines

Front 17

Characteristics of the subject in an experiment or quasi-experiment that the researcher cannot manipulate.

- sometimes used to select subjects into groups.

Back 17

Subject variable

Front 18

If the computed value of R is positive and is also called direct relationship

Back 18

Positive correlation

Front 19

Also called an inverse relationship

Back 19

Negative correlation

Front 20

A non linear trend, range truncation and outlinears.

Back 20

Features of the data

Front 21

Used on statistical formulas for simple correlations, which assumes that the direction of the relationship between X and Y remains the same.

Back 21

General linear model

Front 22

An artificial restriction of the range of values of X or Y

Limited range of data, it could show a range truncation of 0 or close to 0

Outlinears discrupt general linear trend data

Can affect correlational coefficients

Back 22

Range Truncation

Front 23

3 alternative possibilities whenever two behaviors are strongly correlated

Back 23

----

Front 24

inability to reduce cause and effect relationship between two events/variables based on observed association or correlation.

- correlation doesn't imply causation

Back 24

Bidirectional Causation

Front 25

Confounding variable affects both variables to make them seem casually related when they are not.

- third agent may cause two behavior to appear related

Back 25

Third variable problem

Front 26

Estimates the variability in scores one 1 variable that can be explained by other variables.

Estimates strength of relationships between them

Back 26

Coefficient of determination (R2)

Front 27

Argued that r^2 > or equal to 0.25 is considered a strong association between variable.

Back 27

Cohen (1988)