Order #403985 Topic: #3

Chapter_8.pptx

Home >Political Science homework help >Order #403985 Topic: #3

An Applied Reference Guide to Research Designs Quantitative, Qualitative, and Mixed Methods

A Visual System for Research Designs in

Education and the Social & Behavioral Sciences

Edmonds, W. A., & Kennedy , T. D. (2012). An applied reference guide to research designs: Quantitative, qualitative, and mixed methods. Thousand Oaks, CA: Sage.

Part II and Ch. 8

Quantitative methods for Non-experimental research

Sage Publications

Quantitative Methods for non-experimental Research

Method	Quantitative
▼	▼
Research	Non-experimental
▼	▼
Approach	Observational Survey
▼	▼
Design	Posttest (one-group) Ex post facto Explanatory Predictive Cross-sectional Longitudinal

Non-Experimental Research

Most common type of research

Observational data

Intent is not to determine cause-effect relations

3 conditions must to be met in order to determine cause and effect

Covariation?

Temporal Precedence?

No Plausible Alternative Explanation?

Only one form of Control can be utilized for non-experimental research

Manipulation No

Elimination No

Inclusion No

Group or Condition Assignment No

Statistical Procedures Yes

Non-Experimental Research

Statistical Procedures Control

Matching of alternate independent variables (e.g., age, gender, SES)

Causal-modeling

Propensity scoring

Threats to internal validity do not apply

Exception (ex post facto design and the one group posttest only design)

Threats to validity

Construct (e.g., Construct confounding, reactivity to assessment)

External (e.g., sample characteristics)

Statistical Conclusion (e.g., low statistical power)

Unique designs for non-Experimental Research

Ex post facto design

Independent variable has already occurred and is not controlled

Only form of control is statistical procedures

Matched-grouping

Only a post test observation is taken

Although non-experimental, internal validity does apply (e.g., maturation, selection bias)

Group	Treatment	Assignment	Posttest
1	X	MA	O1
2	-	MA	O1
Time ►

Unique designs for non-Experimental Research

One group posttest-only design

Often referred to as the one-shot case study

The independent variable can be controlled via manipulation, inclusion, or elimination.

Only one group and one posttest observation is taken

Although non-experimental, internal validity does apply and there are two major weaknesses associated with this design

Selection bias

Special treatment

Impossible to rule out plausible alternative explanations

Ex post facto design

Research Question Example Here: Fill in the blanks

Formulate a RQ which would necessitate the use of this type of design

Group	Treatment	Assignment	Posttest
1	X	MA	O1
2	-	MA	O1
Time ►

Matched-grouping technique

Group	Treatment	Assignment	Posttest
1		MA
2	-	MA
Time ►

Ex post facto design

Conditions to be met for cause-effect with this design

Covariation?

Temporal Precedence?

No Plausible Alternative Explanation?

Major Threat to internal validity

Selection Bias

Account for lack of group equivalence

Posttest only design (one group)

Research Question Example Here: Fill in the blanks

Group	Treatment	Posttest
1	X	O1
Time ►

Assignment	Group	Treatment	Posttest
NR	1
Time ►

Posttest only design (one group)

Conditions to be met for cause-effect with this design

Covariation?

Temporal Precedence?

No Plausible Alternative Explanation?

Major Threat to internal validity

Selection Bias

Special Treatment

Control techniques to be used with this design

Access companion website to further explore specific references of interest

http://www.sagepub.com/edmonds/

Be cautious of internet resources on research design and methodology.

Be aware that many books use different types of terminology to explain the same thing.

Group 1

Group 2

Group 1

Group 2