Introducing
Your new presentation assistant.
Refine, enhance, and tailor your content, source relevant images, and edit visuals quicker than ever before.
Trending searches
Research question:
Is the “higher education” increasingly approached with a ‘market logic’ over the last 20 years?
(Markets: competition between universities, selling a service; as opposed to public service and good.)
How to approach?
Experiment: not possible
Survey among experts:
- experts do not know either
- likely to give a biased opinion (world-view intervenes)
- people cannot objectively present the past and the future
(e.g. work load among Australian academics)
Solution: analyze the change in the language used in HE strategy documents. -> Use existing government documents!
Method: Content analysis!
1. Find the key HE strategy documents in the last 20 years
"The future of the Higher Education System" (2011) – Veerman committee report
"A strategy plan for the Higher Education: The new millennium" (1999)
…
2. Analyze the content of the documents: coding!
Coding = conceptualization and operationalisation tool
3. Conclusion:
Can you see more and more “market” language in more recent documents?
- Key words: complexity, in-depth
- Not good for descriptive studies that aim to generalize on population: how much, who, and when (survey or document analysis better)
- Good for why and how for theory development appropriate
Manifest content: (higher reliability)
concrete terms present in a communication:
‘Market’, ‘competition’, ‘private benefit’
Latent content: (usually higher validity)
underlying meaning of communications
[often concrete words are not sufficient to convey the true nature]. ‘Tuition fee’ as a market instrument vs social equality instrument. The phrase itself not sufficient.
Validity vs reliability trade-off
Use both if possible!
1. A single case study design: How to cope with conflicting values: The case of desegregation and freedom of choice in Dutch primary education policy
Theory: cycling, firewalling, casuistry
Data: Documents, institutional arrangements, interviews
2. A multiple case study design: How NGDOs reconcile market pressures with social commitment: The case of three NGDOs in the Netherlands (1% club, Terre des Hommes, Cordaid)
Theory: competing institutional logics
Data: mission statements, screening activites, interviews
- Deductive and inductive approach
- Most important: think! What makes an approach market-like or non-market-like
- Different levels: nominal or ordinal
very much market-based (competition)
moderately market-based (profiling)
Not market based (public value)
- What do you analyse:
Sentences
Paragraphs
Stated goals?
Also no apparent orientation important to include. Otherwise a longer document has necessarily a higher score.
Computer programs to help you with the manual work.
- Allows using several data sources
- Theory is crucial, unlike in some other qualitative designs
Collecting your own data:
- survey
- experiment
- qualitative field work
as opposed to
Using existing data:
- Using existing data bases (World Bank data, Eurobarometer, census data) - quantitative research
- Using texts and documents (e.g. legal documents, media texts) - qualitative/quantitative analysis
- Observing what people leave behind:
Ex. What exhibits are most popular at the museum: look where the floor is most worn off.
= you do not “obtrude”, no direct contact with the subjects
Sources:
- newspapers, TV and other media (particulalry good to study values, prejudices, etc.)
- legal documents, government reports
Eur-Lex for European law
http://wetten.overheid.nl Dutch laws
http://www.overheid.nl (Tweede Kamer debates)
- Sector strategy papers
speeches (e.g. prime minister, Queen speech),
- also web content, songs
Sample: cannot include everything
-Stratified sampling (pick the biggest newspapers)
-Quota sampling (pick certain proportion of articles from main national newspapers (Volkskrant), regional (Limburger), free yellow press (Spits)
What is the case? = unit of analysis
a phenomenon: coping with a conflict, in the desegregation example OR organizations: 3 NGDOs as an example
= What is your research question about?
Coping with a value conflict
NGDO’s management issues
Single case studies:
Representative/typical case
Extreme/Unique/critical case
Revelatory case (e.g. excellent opportunity)
Danger: in the end turns out not that representative or unique.
Multiple cases:
Predicts similar results
Predicts different results but for predictable reasons
Construct validity (operationalisation)
Use multiple sources of evidence
Establish key chain of evidence
Have key informants review draft report
Internal validity (causality)
Pattern matching (pattern matches theoretical expectations, or match between multiple cases)
Address rival explanations ( also a form of pattern matching)
…
External validity (generalizable beyond a specific case)
Use theory in a single-case studies
Use replication logic in multiple case studies
Reliability (minimize noise, error)
Use case study protocol
Develop case study database (computer programs)
Exam registration: open until
10 days prior the exam date
Paper draft: due next week Friday
Remember: search methods on Toolbox
www. scholar.google.com
University library: catalogue
Database: EBSCO, JSTOR
Patience!
Strength:
- Possible with not many resources: if you have a source to use, you can complete an interesting high quality study as a student
- Possible to correct errors later in the process. Recoding! (Cannot correct with experiments and surveys)
- Good way to study the past, and processes over long time
- No effect on subjects, unobtrusive
Weaknesses:
- Limited to recorded communications
- Highly valid, if you study communication, but otherwise may be limited validity (biased, part of the data)
E.g. studying prejudices based on media coverage, which is already somewhat self-censured
- Using official or quasi-official data-sets
- We almost always use existing data for background and introduction:
Unemployment and economic growth
Higher education drop-out and completion rate
Regulation intensity
- BUT it can also be main data source for your original analysis
Study a relationship between the feelings of “national pride” and support to the European Union?
International comparative data
OECD Education at a Glance dataset
OECD Economic Outlook
Eurobarometer (Public opinion analysis)
Eurostat
World Bank Health Nutrition and Population data
World Bank Data Catalogue (includes different topics!)
Transparancy International (Corruption and governance)
National data
Centraal Bureau voor Statistiek
Ministries (Education, Economics, Transportation, etc)
...
- Descriptive data presentation
- Statistical techniques (correlation techniques, regression) (M&T2)
- Other (new and cool) techniques
Strength:
- Easily available
- Relatively good quality and trusted (professional and checked)
Validity issues:
- Limited to what we have. Often a measure is not precisely what we want. [“National pride”?]
- How to increase validity:
Logical reasoning (also theory)
Replication: use multiple indicators (angles). If everything refers to the same, then something going on.
Reliability issues:
different definitions over years and countries
e.g. Who is an international student:
Students with a foreign passport (includes immigrants)
Students who received a high-school diploma elsewhere
- Examination of societies (or other social units) over time and in comparison with one another.
- To seek historically grounded answers to large-scale and substantive outcomes
- Searching for patterns (not only presenting one-time historical effect)
PA theses: not encouraged; Commercialisation in public TV
= a synthesis of earlier research
Goal: develop new conclusions
- more observations mean more validity and reliability,
- Helps to specify conditions when something may or may not be the case
Not the same as literature review!
LR presents an overview of the existing evidence, without a goal to suggest a new conclusion
Approach: quantitative (use advanced statistical methods), or
qualitative (synthesize existing case studies)
Public-private cooperation for large infrastructure projects not always a success. Why? When is it and when not?
Koppenjan (2005): studies 9 existing case-studies, some positive and some negative (qualitative)
Looks carefully on the process of the cooperation.
Draws conclusions that certain type of management is essential for success:
- early interaction between partners
- trust between partners
- ...
Koppenjan, J. (2005). The formation of public–private partnerships: Lessons from nine transport infrastructure projects in the Netherlands. Public Administration, 83, 135-157.