Transcript First Two Years Project, St. Lawrence University

First Two Years Project
Cathy Crosby-Currie
Christine Zimmerman
Bringing Theory to Practice
March 2007
Modeling the Multiple Influences
on Civic Development
and Well-Being
Astin’s Theory of Involvement:
I-E-O Model
= How is the output influenced/do students learn
(forces behind something; program/intervention)
Academic or co-curr. program
Engaged Learning Pedagogy
Community Service Program
ENVIRONMENT
B
im
pa
ct
A
C Impact
OUTPUT
INPUT
= Who is learning
(pre-test attributes:
demographics, abilities,
views, etc.)
= What are students learning
(goals and objectives; post-test)
Cognitive
Critical thinking
Academic ability
Voting behavior
Mental Health
Affective
Values
Interests
Satisfaction
Attitudes&Beliefs
Adopted from Astin, A. (1993). Assessment For Excellence: The Philosophy and Practice of Assessment
and Evaluation in Higher Educaiton. Phonenix: ORYX Press, 18
Terenzini’s General Conceptual Model of
College Influence on Student Learning
Terenzini, P., Springer, L., Pascarella, E., Nora, A.(1995). Influences Affecting the Development of
Students’ Critical Thinking Skills. Research in Higher Education, 36, 23-39.
Methodological
Considerations
Experimental v. Quasi-Experimental
Design

Key difference between experimental
and quasi-experimental

Researcher’s control over the “input”
variable



Experimental: YES! -> cause/effect conclusions
Quasi-Experimental: NO! -> examine
relationships
Control v. comparison groups
Experimental v. Quasi-Experimental
Design

Quasi-experimental power comes
from:

Ability to detect change through design


e.g., interrupted time series design
Equivalence of comparison group to
experimental group
Longitudinal v. Cross-Sectional
Designs
 Cross-sectional:
 Comparing
groups of different
ages at one point in time
 Convenient but lacks statistical and
conceptual power
 Longitudinal:
 Comparing
individuals to
themselves across time
 Multiple cohorts is ideal
St. Lawrence’s Quasi-Experimental,
Longitudinal Design
 Participants:
 Two
Cohorts – Selected Students
from Classes ‘09 and ’10
 Experimental Group – students in
Brown College
 Second year added a second
experimental group
 Comparison Group – non-equivalent
group matched on key variables of
interest
 Comparison Group Sample II.xls
St. Lawrence’s Quasi-Experimental,
Longitudinal Design

Data Collection
 Pretest (9/05 & 9/06)
 Posttest (2/06 & 2/07)
 Follow-up (4/07 & 4/08)
Challenges of Quasi-Experimental
and/or Longitudinal Designs
 Creating
comparison group(s)
 Participant attrition
 Communication
incl. letter from
president
 Personalized letters & email
 Contacting students multiple
times/multiple ways
 Accommodate students’ schedules

Institutional Review Board Approval
 Reframe
as a positive contribution to
your research not a hurdle to overcome
Challenges of
Measurement
• Valid and Reliable Measures
• Direct - Indirect Measures
• Quantitative - Qualitative Data
• Process - Outcomes Measures
Reliability
… is the consistency or repeatability
of responses

Random error (noise)

Systematic error (bias)
Reliability (cont.)
Ways to increase data reliability:






Clear directions
Clear questions
Consistent order of questions
Clear survey layout
Trained proctors/interviewers
Consistent data entry and scoring
Reliability (cont.)
How to assess the reliability of your
instrument:

Pilot-test your study

Test-retest your survey

Focus-group survey or interview questions

Include similar questions in same
questionnaire
Validity
… the extent to which the instrument
truthfully measures what we want to
measure

How well does the instrument content match what
we want to measure?

Do respondents interpret the questions correctly?

Do respondents’ answers reflect what they think?

Are the inferences we make from this study
accurate? Can they be generalized?
Validity (cont.)
How to establish validity:

Use multiple measures and multiple methods

Derive measures from literature review &
existing research / participate in national
survey instruments and tests

Expert review

Pilot-test your own survey
Direct – Indirect Measures
Direct: tangible, actual evidence
Indirect: proxy for what we try to measure
Direct Measures
 Portfolio
Indirect Measures
 Self-reported behavior,
attitudes, gains

Essay/reflection

Performance task/test

Grades

Actual student behavior

Participation rates

Time spent at task
Qualitative – Quantitative Measures
Qualitative: unit of data = words
Quantitative: unit of data = numbers
Qualitative Measures
 Focus groups
 Structured interviews
 Self-reflections/diaries
 Open-ended survey
questions
Quantitative Measures
 Surveys with closed
questions (Likert scale,
check list, etc.)
 Grades
 Actuary data such as
participation rates,
attendance, etc.
Process – Outcomes Measures
Process Measures

What did we do?
(=data to demonstrate the implementation
of an activity/program)
Outcomes Measures

What are the results?
(= data used to measure the achievement
of an objective/goal)
 Initial
 Intermediate
 Long term
Administrative Challenges
And Best Practices

Buy-in and Support


Form campus partnerships early on
Build on existing data collections

Institutional survey cycles and survey
timing

Copyrights of survey instruments

Liability for use of certain measures

Survey recruitment & retention
Select Survey Instruments
and Literature
Sampling of Survey Instruments

Entering Student Survey



CIRP Freshman Survey (HERI, UCLA)
College Students Expectations Questionnaire
CSXQ (Indiana)
Enrolled Undergraduate Students/Alumni




As a continuation of CIRP: Your First College
Year/College Senior Survey
As a continuation of CSXQ: College Student
Experience Questionnaire (CSEQ)
National Survey of Student Engagement (NSSE)
Consortia Senior and Alumni Surveys (e.g. HEDS.
COFHE)
Sampling of Survey Instruments

Depression/Mental Health Measures



Optimism/Pessimism/Happiness Scales



Beck’s Depression Inventory (BDI II)
Brief Symptoms Inventory (BSI)
Mehrabian Optimism/Pessimism Scale
http://www.authentichappiness.sas.upenn.edu/
Alcohol/Drugs/General Wellness


CORE Alcohol And Other Drugs Survey
ACHA-NCHA
Sampling of Survey Instruments

Civic Development (from Lynn Swaner)


Other National Surveys


CASA TELEPHONE SURVEY INSTRUMENT
HERI Faculty Survey
Other In-House Institutional Surveys



Course evaluations
Program evaluations
Satisfaction studies
Select Literature

The National Center on Addiction and Substance Abuse at
Columbia University (2003): Depression, Substance Abuse, and
College Student Engagement: A Review of the Literature. Report
to The Charles Engelhard Foundation and The Bringing Theory to
Practice Planning Group.
http://www.aacu.org/bringing_theory/research.cfm

The National Center on Addiction and Substance Abuse at
Columbia University (2005): Substance Abuse, Mental Health and
Engaged Learning: Summary of Findings from CASA’s Focus
Groups and National Survey. Report to Sally Engelhard Pingree
and The Charles Engelhard Foundation for the Bringing Theory to
Practice Project, in partnership with the Association of American
Colleges and Universities.
http://www.aacu.org/bringing_theory/research.cfm

Swaner, L.E. (2005). Linking Engaged Learning, Student Mental
Health and Well-being, and Civic Development: A Review of the
Literature. Prepared for BTtoP
http://www.aacu.org/bringing_theory/research.cfm
Select Literature

Pascarella, E., Terenzini, P.(1991). How College
Affects Students: Findings and Insights from
Twenty Years of Research. San Francisco:
Jossey-Bass.

Bringle, R. G., Phillips, M.A., Hudson, M. (2004).
The measure of service learning: Research
scales to assess student experiences.
Washington, D.C. American Psychological
Association

Suskie, L. (1996). Questionnaire Survey
Research: What works. Tallahassee: Association
for Institutional Research