Module 5

📘 Chapter 9: Enhancing the Validity of Research

✅ Types of Validity

• Internal Validity: Confidence that the intervention caused the observed effect.

• External Validity: Generalizability of the results to other populations or settings.

⚠ Threats to Internal Validity

1. Historical effects: Outside events affecting results.

2. Maturation effects: Natural changes in participants.

3. Testing effects: Improvement due to repeated testing.

4. Instrumentation effects: Inconsistent data collection tools/methods.

5. Consent effects: Differences between those who consent and those who don’t.

6. Treatment effects (Hawthorne effect): Behavior changes due to being studied.

7. Multiple-treatment effects: Overlap of multiple interventions.

8. Selection effects: Non-random group assignments.

9. Attrition: Participant drop-out affects results.

🛡 Minimizing Threats to Internal Validity

• Match design to research question.

• Control for bias in:

  • Selection

  • Data collection

  • Analysis

  • Publication

• Use proper statistical analysis (e.g., check for Type I/II error).

🌍 External Validity

• Population Validity: Can findings apply to a larger group?

• Ecological Validity: Can findings apply to different environments?

🔺 Jeopardizing Factors

• Selection bias: Sample doesn’t reflect population.

• Time/history effects: Changes over time or unusual events.

• Novelty effect: Participants react to the "newness" of intervention.

• Experimenter effect: Participants react to the researcher.

⚖ Balancing Internal & External Validity

• High internal validity = More control = Less generalizability.

• Balance is needed to produce useful and applicable results.

🧠 Trustworthiness in Qualitative Research

According to Schou et al. (2012):

• Credibility

• Transferability

• Dependability

• Confirmability

Amin et al. (2020) add:

• Plausibility

• Believability

• Applicability

✔ Strategies to Promote Qualitative Validity

• Prolonged fieldwork

• Verbatim accounts & triangulation

• Member checking (participant feedback)

• Bracketing (acknowledge researcher bias)

• Audit trails (document decisions & data)

🔍 How to Assess Validity in a Study

• No specific “validity” section in most studies.

• Look at:

  • Methods and Procedures

  • Discussion/Conclusion

  • Signs of replication

📘 Chapter 11: Summarizing & Reporting Descriptive Data

🧮 Purpose of Descriptive Data Analysis

• Provides summary of data

• Types:

  • Counts

  • Central tendency (mean, median, mode)

  • Variability (range, SD)

  • Position (percentiles, z-scores)

  • Relationships (correlations)

  • Graphical presentations

📊 Levels of Measurement

🔢 Frequencies & Distributions

• Frequency = how often something occurs

• Tables: Sort and count occurrences

• Graphs: Bar charts, histograms, etc.

🔺 Distributions

• Normal: Bell-shaped

• Skewed: Left/right lean

• Kurtosis: Peaked or flat shape

🎯 Measures of Central Tendency

📉 Measures of Variability

• Range: High – Low

• Variance: Spread from the mean

• Standard Deviation (SD): Square root of variance

📍 Measures of Position

• Percentiles: % of scores below a value

• z-scores: Distance from mean in SD units

🔗 Measures of Relationship

• Correlation:

  • Positive or negative

  • Strength & direction via correlation coefficient

• Graphical tools: Scatter plots

📈 Graphical Presentations

❌ Common Errors

• Using wrong statistics

• Interpreting data out of context

• Overstating results

• Misrepresenting data visually

🧐 Reading Descriptive Data in Studies

• Understand symbols (sample = x̄, population = μ)

• Match analysis with measurement level

• Assess relevance and appropriateness

💡 Using Descriptive Data in Practice

• Foundation for intervention design

• Provides insights into patterns and trends

📝 How to Report Descriptive Data