Descriptive Research (I) Week 8 Lecture 1
Thursday, May. 6, 2004
1
Agenda • Quantitative Design issues • Correlation Research • Survey Research
Thursday, May. 6, 2004
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
2
Quantitative design issues • The language of variables and hypotheses – Quantitative researchers redefine concepts of interest into the language of variables – Attributes: the values or the categories of a variable – Examples: • Male, married, number of years married • SSL, SET, PKI, security technology
– Types of variables • Independent variable • Dependent variable
Thursday, May. 6, 2004
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
Introduction/What is research?
3
1
Quantitative design Issues • The language of variables and hypotheses – An expected but unconfirmed relationship between two or more variables – Where do hypotheses come from • Theory, direct observation,guess, intuition
– Two different ways of stating a hypothesis • Looking for difference between variables • Looking for relationship between variables
Thursday, May. 6, 2004
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
4
Refinement of hypothesis • Problem or general hypothesis – You expect some children to read better than others because they come from homes in which there are positive values and attitudes to education.
• Research hypothesis – Reading ability in nine-year-old children is related to parental attitudes towards education
• Operational hypothesis – There is a significant relationship between reading ability for nine-year-old children as measured by standardized reading test X and parental attitudes to education as measured by attitude test Y Thursday, May. 6, 2004
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
5
What is descriptive research • Identify the characteristics of an observed phenomenon • Explore possible correlations among two or more phenomena
Thursday, May. 6, 2004
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
Introduction/What is research?
6
2
Correlational Research • Examines the extent to which differences in one characteristic or variable are related to differences in one or more other characteristics or variables. • Gather data about two characteristics for a particular group of people or other appropriate units of study.
Thursday, May. 6, 2004
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
7
Example of correlational research
Scattergram: dynamic relationship between age and reading level Thursday, May. 6, 2004
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
8
Interpretation • Describe the homogeneity or heterogeneity of the two variables • Describe the degree to which the two variables are intercorrelated • Interpret these data and give them meaning.
Thursday, May. 6, 2004
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
Introduction/What is research?
9
3
Cautions about interpreting correlational results
• Correlation does not, in and of itself indicate causation • One variable correlates meaningfully with another only when a common causal bond links the phenomena of both variables in a logical relationship
Thursday, May. 6, 2004
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
10
Survey research • General features – Large-scale probability sampling • A study on snipping behavior of online auction took 368 eBay bidders as respondents
– Systematic Procedures: Interviews and questionnaires – Answers are numerically coded and analyzed with the aid of statistical software – Used extensively for both descriptive and explanatory purposes Thursday, May. 6, 2004
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
11
Research questions appropriate for a survey • Developed within the positivist approach to social science • Self-reported beliefs or behaviors. • Ask many things, measure many variables and test several hypotheses in a single survey – – – – –
Behavior Attitudes/beliefs/opinions Expectations Self-classification Knowledge
Thursday, May. 6, 2004
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
Introduction/What is research?
12
4
Steps in conduction a survey •
Step 1 Questionnaire design – Develop hypotheses – Decide on type of survey (mail, interview, telephone) – Write survey questions – Decide on response categories – Design layout
•
•
– Locate respondents – Conduct interviews (distribute questionnaire) – Carefully record data
•
Step 3 Sampling – – – –
Decide on target population Get sampling frame Decide on sample size Select sample
Step 5 Data analysis – Enter data into computers – Recheck all data – Perform statistical analysis on data
Step 2 Pilot test – Plan how to record data – Pilot test survey instrument
•
Step 4 Data collection
•
Step 6 Writing up – Describe methods and findings in research report – Present findings to others for critique and evaluation
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
Thursday, May. 6, 2004
13
Types of surveys • Self-administrated Questionnaires – Hand delivered – Mail survey – Web survey
• Face-to-face and Telephone interviews – Structured interview • The researcher asks a standard set of questions and nothing more
– Semi-structured interview • The researcher may follow the standard questions with one or more individually tailored questions to get clarification or probe a person’s reasoning
Thursday, May. 6, 2004
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
14
Questionnaire item styles • Open-ended question – Respondents answer in their own words
• Closed-ended question – Respondents choose a response from those provided
• Examples – What drives you to bid in the last minutes in a second-price auction? (OPEN) – Which one of the factors listed below drives you to bid in the last minutes? (CLOSE) • () Checking around if there are other auctions offering the same item • () Don’t want other bidders get benefit from my expertise of the auction item • () Avoid bidding war • () Other _____________(please explain) Thursday, May. 6, 2004
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
Introduction/What is research?
15
5
Open vs. closed • Open-ended question – Great freedom for respondent to answer – Responses may be ambiguous – Coding is time-consuming and costly which usually results in some degree of error – Entail more work from respondents
• Closed-ended question – Require less effort and less facility with words – Difficult to develop good closed questions – Recommendation for designing closed-end question: use open questions in preliminary interviews or pretests ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
Thursday, May. 6, 2004
16
Ranking and scaling • Rank-ordering questions Rank-order the most important things you want in the job you make your life’s work? (1 indicates the most important one 4 the least important one) ______
Making a lot of money
______
Being creative
______
Being free from supervision
______
Having opportunities for advancement
• Rating scale questions – Respondents are asked to indicate the degree of their agreement or disagreement with a statement What’s your overall reaction to the present Bear Lake bus system 1
2
3
4
5
Very satisfied
Satisfied
Neutral
Dissatisfied
Very dissatisfied
Thursday, May. 6, 2004
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
17
Using language effectively • Are the items unambiguous, easily read and sufficiently brief – Avoid indefinite words such as “usually”, “seldom”, “many”, “few”, “here”, “there” How often do you use Internet? (bad example) ()
Seldom or never
()
Often
()
Very often
• Avoid false premises The post office is open too many hours. Do you want it to open four hours later or close four hours earlier each day? (bad example) ()
Open four hours later
()
Close four hours early
Thursday, May. 6, 2004
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
Introduction/What is research?
18
6
Using language effectively(II) • Avoid double-barreled question – two separate ideas are presented together as a unit • The system is easy to use and helpful
• Avoid overlapping or unbalanced response categories.
Thursday, May. 6, 2004
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
19
Questionnaire design issues • Length of survey or questionnaire • Question order or sequence – Opening, middle and ending questions – Group questions on the same topic together. – Order effect
• Format and Layout • Noresponse – The percentage of people who have and have not consented to participate • Those who agreed or refused to be interviewed • Those who have or have not returned questionnaire Thursday, May. 6, 2004
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
20
Question order effect Question Order Effects Question 1 “Do you think that the United States should let communist newspaper reporters from other countries come in here and send back to their papers the news as they see it? Question 2 “Do you think a Communist country like Russia should let American newspaper reporters come in and send back to America the news as they see it?” PERCENTAGE SAYING YES Heard first
Yes to #1 (Communist Report)
Yes to #2 (American Reporter)
Question 1
54%
75%
Question 2
64%
82%
Thursday, May. 6, 2004
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
Introduction/What is research?
21
7
Analyzing one variable • Univariate frequency distribution Degree of satisfaction Very satisfied
Very disatisfied
Frequency
Percent
Cumulative percent
1
23
16.4
16.4
2
71
50.7
67.1
3
19
13.6
80.7
4
14
10.0
90.7
5
11
7.9
98.6
2
1.4
100.0
140
100
1
2
3
4
5
6
19
6
Total
Thursday, May. 6, 2004
14 71
11 2 23
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
22
Being critique on survey results • The items to include when reporting survey research – – – – – – – – – –
The sampling frame used The date on which the survey was conducted The population that the sample represents The size of the sample for which information was collected The sampling methods The exact wording of the questions asked The method of the survey The organizations that sponsored the survey The response rate Any missing information or “don’t know” responses when results on specific questions are reported
Thursday, May. 6, 2004
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
23
Summary • Quantitative design issues – Language of variables and hypotheses
• Correlational research – Scattergram and it’s interpretation
• Survey research – Steps – Different types of survey – Design of questionnaire – Preliminary analysis of survey data Thursday, May. 6, 2004
ISYS3015 Analytical Methods for IS Professionals School of IT, The University of Sydney
Introduction/What is research?
24
8
