Data Analysis in Quantitative and Qualitative Research Methods
Data Analysis in Research
- Data analysis depends on research design, questions, and hypothesis.
- Data can be quantitative (numerical), qualitative (verbal), or mixed.
- Data analysis describes participant responses, noting typical or unusual aspects, differences, relationships, and answers to research questions.
Quantitative Data Analysis
- Research verifies ideas and theories by gathering information to answer questions.
- Statistical analysis is used when numbers represent information.
- Statistics are mathematical techniques to examine data and test theories.
- Effective statistics organize, evaluate, and analyze data to answer research problems.
- Two classes of statistical techniques: descriptive and inferential.
Descriptive Statistics
- Descriptive statistics summarize and describe collected data.
- Research results are represented as percentages, proportions, ratios, and rates.
- Proportion: P = f/n
- Percentage: % = (f/n) \times 100
- Where:
- f = frequency
- n = total number of cases
- Ratios compare the number of cases in categories of a variable.
Measures of Central Tendency
- Numerical values that locate the center of data: mean, median, mode, and midrange.
- Mean ((\bar{x})): Sum of all values divided by the number of samples.
- Formula: \bar{x} = \frac{\Sigma x}{n}
- Median: Middle value when data is ranked.
- Depth of median: d(x) = \frac{\text{sample size} + 1}{2}
- Mode: Most frequent value in a data set. If no number occurs more than once, the sample has no mode.
- Midrange: Number midway between the lowest (L) and highest (H) values.
Measures of Dispersion
- Analyze data spread or variability: range, variance, and standard deviation.
- Range: Difference between the highest (H) and lowest (L) values.
- Sample Variance (s^2):
- Formula: s^2 = \frac{\Sigma(x-\bar{x})^2}{n-1}
- Standard Deviation (s):
- Square root of the variance: s = \sqrt{s^2}
Inferential Statistics
- Uses sample data to infer about the sampled population.
- Inferences include estimating population parameters and testing hypotheses.
- Null Hypothesis (H_0): Hypothesis being tested.
- Alternative Hypothesis (H_1): Research hypothesis.
- Decision Rule:
- If p-value ≤ level of significance (e.g., 0.05), reject H_0.
- If p-value > level of significance, fail to reject H_0.
- Correlation: Measures the relationship between two variables.
- Pearson's r: Describes the relationship between two continuous variables.
Qualitative Data Analysis
- Based on logic and observation.
- Qualitative data: information in words rather than numbers.
- Focus on meanings, context, and experience.
- Values subjectivity, analyzed with philosophical assumptions.
- Goal is to understand meanings and how people make meaning.
Grounded Theory Analysis Tasks
- Researcher prepares verbatim transcripts
- Anonymize data
- Develop codes
- Define codes in a codebook
- Code data
- Describe
- Compare
- Categorize
- Conceptualize
- Develop theory
Data Preparation for Qualitative Analysis
- Verbatim transcription of interviews or discussions.
- Translation of transcripts, if needed.
- Removal of identifiers to ensure participant anonymity.
Developing Codes
- Codes represent issues, topics, ideas, or opinions evident in the data.
- Inductive codes: Raised by participants.
- Deductive codes: Prompted by the interviewer.
Making a Codebook
- Provides a central reference for all codes.
- Each code has a name and definition.
Interpretation of Qualitative and Quantitative Data
- Basic research develops reliable knowledge.
- Data should be reduced into smaller units or categories.
- Appropriate analysis method depends on the research approach.
Interpreting Quantitative Data
- Focuses on explaining numerical data
- Uses a deductive method for theory and hypothesis testing.
- Variables: nominal, interval, ordinal, and ratio.
Interpreting Qualitative Data
- Data are non-numerical (words or pictures).
- Data is reduced to patterns, categories, or themes.
- Uses an inductive method to explore phenomena.
- Thematic analysis: Segmentation, categorization, and linking of data aspects before final interpretation.