Transcript
Page 1: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570Session 5.1

Research Design and Data Collection

Page 2: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 2

Objectives• Place sampling and research

design in context • Recall components of research

design• Further develop the research

projects• Get feedback on research designs

Page 3: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 3

Agenda• Research in context• Review of research design

– Definition– Criteria for selection– Research goal (type): explanatory;

descriptive• Common designs• Exercise: “elevator speech”• Exercise: developing questions

Page 4: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 4

Sampling and Research

• Reality, models, & research findings• Absolutes, certainty, beliefs, & decisions • Sampling: done if a census is not

feasible because of resources or other constraints

• Point of research (including sampling):– Building knowledge through

• communication with others• storing and making accessible results

– Following “accepted” methods

Page 5: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 5

Research Design: Definition

• Framework which guides the collection of data

• Roadmap or blueprint for the study• Goal: select the strongest (or most

appropriate) design for the type of research

• Two major types of research– descriptive– explanatory

Page 6: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 6

Choice of research design (Bouma)

Depends on hypothesis or research objective • Case study: What is happening?• Longitudinal study: Has there been a

change in A• Comparison: Are A and B different?• Longitudinal comparison: Do A and B differ

over time?• Experiment: Is the difference between A

and B due to a change in the independent variable?

Page 7: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 7

Case study• Simplest (most primitive?) design

– Rich in detail– Requires “plausibility” to establish

generality– Documentation/write up is not trivial

• Collecting information from one group at one point in time

• Aim: description—what is going on? Are there discernable relationships among the factors?

Page 8: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 8

The longitudinal study

• Two or more case studies of the same group separated by an interval of time

• Measure a variable at both points of time

• Intervention: cannot conclude a causal influence

InterventionTime 1 Time 2

Page 9: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 9

Comparison study• Cross-sectional design• Most common• Collect measures from at least two

groups of people at one point in time

• Compares the extent to which the two groups differ on the dependent variable

Page 10: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 10

Comparison Study• Select variable related to the concept

under study• Devise a way to measure the variables• Develop a data recording method• Measure the same variable: use same

technique, measure at the same/nearly the same timeA

B

Do A and B differ?Time 1

Page 11: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 11

Longitudinal comparison

• Study different groups at two points of time

• Can avoid problems of keeping track of individuals over time

• Problems– Can’t draw causal inferences– Matching the samples at time 1and time 2

Page 12: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 12

Explanatory Research: Basic Requirements

• A relationship between the variables needs to be established

• All other reasons for the relationship must be able to be eliminated

• Strongest research design is an experimental design– Vary independent variable– Control for other variables

Page 13: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 13

Classic Experimental Designs: Minimum

Requirements• Two groups– Experimental group – Control group

• Random Assignment of units of analysis to C and E groups– Active intervention for the E group

• Purposefully change the value of the Independent Variable

• Synonyms (other “jargon” in the literature): “treatment,” “manipulation of the IV”

– Measure dependent variable after intervention for E and C groups

Page 14: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 14

Collecting data• How the data will be collected

– Observation– In-depth interview (structured, semi-

structured)– Content analysis

• Questionnaire• Which questions to ask?

– Research problem– Indicators– Hunches about the link between variables

Page 15: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 15

Questionnaires• Explanatory research

– Measures of the dependent variable– Measures of the independent variable– Measures of the test variable– Background (measures of control variables

—ceteris paribus)• Descriptive research

– The phenomenon we are describing– Background characteristics

Page 16: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 16

Question content• Behavior: what people do• Beliefs: what people believe is

true or false• Attitudes: what people think is

desirable• Attributes: characteristics

Page 17: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 17

Wording questions (De Vaus, Babbie)

• Is the language simple?• Can the question be shortened?• Is the question double-barrelled

(compound)?• Is the question leading?• Is the question negative?• Is the respondent likely to have the

necessary knowledge?• Will the words have the same meaning for

everyone?• Is there a prestige bias?

Page 18: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 18

Wording questions• Is the question ambiguous?• Do you need a direct or indirect

question?• Is the frame of reference for the

question sufficiently clear?• Does the question artificially create

opinions?• Is personal or impersonal wording

preferable?

Page 19: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 19

Wording questions• Is the question wording

unnecessarily detailed or objectionable?

• Does the question have dangling alternatives?

• Is the question likely to produce a response set?

Page 20: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 20

Question types (De Vaus)

Open or closed questions?• Open Questions: how to code

responses• Closed question formats

– Category or rating scales– Semantic differential formats– Checklists– Ranking formats– Attitude choice

Page 21: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 21

SummarySimple case study

Longitudinal study

Comparison

Longitudinalcomparison

Experiment

What is happening?

Has there been a change in A

Are A and B Different?

Are A and B different through time?

Is the difference between A and B due to a change in the independent variable

Page 22: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 22

Elevator Speech20-30 second synopsis; intent: to

elicit interest• Who you are and what you are

doing• With whom• Where/How• Why: What you hope to find, why

the results may be important

Page 23: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 23

Exercise• In teams, prepare your “elevator

speech”• Practice so that everyone can

deliver it

5 minutes

Page 24: The Information School of the University of Washington LIS 570 Session 5.1 Research Design and Data Collection

The

Info

rmat

ion

Scho

ol o

f the

Uni

vers

ity o

f W

ashi

ngto

n

LIS 570 Research Design Mason; p. 24

Exercise• Discerning problems in question

construction• In teams, review the handout

questions—how can they be improved?


Recommended