Analysis of Research Findings Flashcards
1
Q
Process of Quantitative Research (11 steps)
A
- Theory
- Hypothesis
- Research design
- Devise measures of concepts
- Select research sites
- Select research subjects
- Collect data/administer research instruments
- Process data
- Analyze data
- Findings/conclusions
- Write up findings/conclusions
2
Q
Analysis Process
A
- Typical amount of data is huge (1000-1500 rows/items)
- Variables can be 200-250 columns
- Must focus your analysis on specific issues
- Hypothesis is used to design questions and variables; and helps again during analysis
- Focus on answering your hypothesis
- Exploration is useful and important, but you don’t need to write or describe each variable as they appear in your data
- Research is not just informing, but you should also provide some sort of solution to the problem; research should serve as a tool for decision-making
- Give recommendations or interpretations; that’s often what clients are paying for
- To have the evidence is one thing but expectations are higher
- You should provide your own input
- You should present, show, and then distinguish your interpretation as part of a second section; could be subjective
- Is about the search for explanation and understanding, in the course of which concepts and theories are likely to be advanced, considered and developed
- When you analyze something you examine it in detail in order to discover meaning or to discover and define its essential features
- A good analysis makes good research
- Two people may see two different analyses from the same data
- Process of analysis:
- Described in some books as a different domain; not like that in the real world
- You do analytical work together with writing the final report
- Important to be reflexive and agile
3
Q
Principles of Analysis (4 elements)
A
- Data/Information
* * What? Who? Where? - Scientific reasoning/argument
* * What happens? How? - Finding
* * What results? - Lesson/conclusion
* * So what? So how? Therefore…
* ** Often if you read reports it’s only on the level of the data, but that provides no real conclusion. What’s the reason? To what extent is this data significant?
* ** How should your client’s decision differ/change based on these findings?
* ** Must use statistical tests
* * Must define your analytical framework; often for yourself; draw a story
* ** Helps you to be coherent when you describe the data
* ** At the beginning of your report, you may argue something and by the end of your research
* ** Helps you be consistent in your arguments/conclusions
* ** Readers/decision-makers expect some sort of straight-forward recommendations
4
Q
Main Elements of Analysis (6 items)
A
- Comprehending
- Explanation
- Synthesizing
- Theorizing
- Recontextualizing
- Interpretation
5
Q
Element of Analysis
1. Comprehending
A
- Full understanding of the setting, culture, and study topic before research begins
- Must understand the context and social reality
- Not to prove something or explain, but to have a full understanding
6
Q
Element of Analysis
2. Explanation
A
- Explanations are the statements which make something intelligible about why things are the way they are
- How do two variables relate to each other
- What’s behind it?
- What’s the pattern, trend, or causal relationship?
7
Q
Element of Analysis
3. Synthesizing
A
- Drawing together of different themes from the research and forming them into new integrated patterns
- If you lack analytical framework, it’s very difficult to do a synthesis of the information
- Must be able to distinguish what’s important and what’s not
8
Q
Element of Analysis
4. Theorizing
A
- Constant development and manipulation of malleable theoretical schemes until the ‘best’ theoretical scheme is developed
- “There’s nothing more practical than a good theory”
- To what extent does your research fit with your theoretical model
- Don’t have to explain all elements of your data, just use theories
9
Q
Element of Analysis
5. Recontextualizing
A
- Process of generalization so that the emerging theory can be applied to other settings and populations
- Putting info in context might be a good strategy to show the nature of your findings
- Can show a certain pattern within one subgroup
10
Q
Element of Analysis
6. Interpretation
A
- Helps the reader to make sense of the data; look for empirical assertations supported by the data
- Key process of analysis; most of what we do
- Helping people find meaning in the data
11
Q
Principles of Interpretation
A
Basic guide to data interpretation:
- “Analyze”, not “narrate”
- Break down into research objectives and research questions
- Identify phenomena to be investigated
- Virtualize the “expected” answers and validate the answers with data
- Don’t tell something not supported by data
When analyzing:
- Be objective (as much as possible)
- Accurate
- True
Separate facts and opinions:
- Do not ignore the facts, especially if they show something different from what you expect
- How you might notice predictors of behavior, something special or different
- You may need to redesign the way you plan to write your report, or you may need more time for analysis, but it’s still good to do it because that is what will give sense/meaning to all your research efforts
12
Q
Avoiding Analysis Mistakes
A
- Read literature on data analysis techniques
- Get a better idea of what fits your goals best
- Evaluate various techniques that can do similar things with respect to the research problem
- Results may differ for different people
- Many international comparative studies use various techniques
- Guidelines of how to be reflective
- Know what a technique does and what it doesn’t
- Consult people; especially your supervisors (or professors)
- Pilot-run the data and evaluate results
13
Q
Presenting Numerical Information (8 types)
A
Only thing your clients will “see” from the research. Only tangible part of the research.
- Frequency tables
- Simple statistics
- Simple bar charts
- Cross tabulations
- Comparative charts
- Scatterplot
- Time series data
- Time series charts
14
Q
Frequency tables
A
In this example, the number of errors cited when 3 different websites are tabulated
* Present findings in percentages; standard practice for research
Columns: Website A, B, C
Rows: Errors reported; percentage
15
Q
Simple statistics
A
Data needs to be summarized using a variety of standard tools:
- Proportions, expressed in percentages
- Means, medians, etc.
- Measures of variability, such as ranges, standard deviations, etc.
Columns: Expert, Website A, B, C
Rows: I, II, III, IV, V, VI, Mean, St. Dev.
- If you present mean, you must also include standard deviation
- Might have satisfaction score in the middle of the scale but two situations
- U-shape indicates two different subgroups (one satisfied, one unsatisfied) although mean might bein the middle
