Data Collection and Analysis

Essential parts of research · gather information, process it, draw valid conclusions

1. Method Validation (Acceptance of Method Validation)

Definition: Method validation is the process of proving that a research method, analytical procedure, or experimental technique consistently produces accurate and reliable results.

Validation ensures that: the method measures what it is intended to measure; results are reproducible; data generated are trustworthy.

Objectives of Method Validation

• Ensure accuracy
• Ensure precision
• Confirm reliability
• Improve reproducibility
• Meet scientific and regulatory requirements

Parameters of Method Validation

• a. Accuracy – measures closeness between observed and true values.
  Accuracy = (Measured Value / True Value) × 100
  Example: True concentration = 100 mg, Measured = 98 mg → Accuracy = 98%
• b. Precision – consistency of repeated measurements. Types: Repeatability, Intermediate precision, Reproducibility. Measured using SD, %RSD.
  RSD = (SD / Mean) × 100
• c. Specificity – ability to measure only the intended parameter without interference. Example: Drug analysis without interference from impurities.
• d. Linearity – ability to produce results proportional to concentration. Example: Concentration increases → instrument response increases. Measured by R² ≈ 1.
• e. Range – interval between lowest and highest concentrations.
• f. Detection Limit (LOD) – smallest quantity detectable. LOD = 3.3 × (σ / S)
• g. Quantification Limit (LOQ) – lowest quantity measurable accurately. LOQ = 10 × (σ / S)

2. Observation and Collection of Data

Observation means systematically watching and recording events, behavior, or measurements.

Types of Observation

• Direct Observation – researcher observes directly (e.g. watching patient behavior).
• Indirect Observation – data obtained through records (e.g. hospital reports).
• Participant Observation – researcher becomes part of study group.
• Non-Participant Observation – researcher observes without participating.

Data Collection – gathering information for research. Objectives: obtain accurate information, answer research questions, test hypotheses.

3. Methods of Data Collection

A. Primary Data Collection

• a. Survey – collect data using questionnaires. Advantages: large sample, cost-effective. Disadvantages: response bias.
• b. Interview – types: structured, semi-structured, unstructured. Advantages: detailed information.
• c. Observation – recording actual behavior.
• d. Experiments – controlled conditions to test variables (e.g. drug effectiveness study).
• e. Focus Groups – group discussion for opinions.

B. Secondary Data Collection

Already available data. Sources: journals, government reports, books, databases. Advantages: fast, economical. Limitations: may be outdated.

4. Sampling Methods

Definition: Sampling is selecting a subset from a population. Population → entire group; Sample → selected participants.

A. Probability Sampling

• a. Simple Random Sampling – random selection (e.g. lottery method).
• b. Systematic Sampling – select every nth participant (e.g. every 10th patient).
• c. Stratified Sampling – divide population into groups (e.g. male/female groups).
• d. Cluster Sampling – select entire groups (e.g. select one school from district).

B. Non-Probability Sampling

• a. Convenience Sampling – easy availability.
• b. Purposive Sampling – selected intentionally.
• c. Snowball Sampling – participants recruit others.
• d. Quota Sampling – fixed numbers per category.

5. Data Processing

Definition: Converting raw data into meaningful information.

Steps

• a. Editing – check errors and completeness.
• b. Coding – assign numbers/symbols (e.g. Male = 1, Female = 2).
• c. Classification – group similar data.
• d. Tabulation – arrange data into tables. Example:

Age Group	Frequency
18–25	45
26–35	30

• e. Data Cleaning – remove duplicates and errors.

6. Data Analysis Strategies and Tools

Definition: Process of examining data to discover patterns and conclusions.

Types of Data Analysis

• A. Descriptive Analysis – summarizes data. Measures: Mean, Median, Mode, Standard deviation. Example: Average age of participants.
• B. Inferential Analysis – draw conclusions from sample. Methods: hypothesis testing, confidence intervals, regression.
• C. Exploratory Analysis – identify patterns.
• D. Predictive Analysis – forecast future outcomes.

7. Data Analysis Using Statistical Packages

Statistical software helps analyze data efficiently.

A. SigmaSTAT

Purpose: statistical analysis, graph generation. Functions: t-test, ANOVA, regression, correlation. Advantages: user-friendly, scientific applications.

B. SPSS (Statistical Package for Social Sciences)

Applications: data entry, statistical analysis, visualization.

Steps in SPSS:

1. Enter data
2. Define variables
3. Select analysis
4. Run statistical test
5. Interpret output

8. Student’s t-Test

Definition: Compares means of two groups.

Types

• a. Independent t-test – two different groups (e.g. Drug A vs Drug B). t = (x̄₁ – x̄₂) / SE
• b. Paired t-test – same group before and after treatment.

Interpretation: p < 0.05 → Significant; p > 0.05 → Not significant.

9. ANOVA (Analysis of Variance)

Definition: Compares means among three or more groups.

Types

• a. One-Way ANOVA – one independent variable (e.g. compare three medicines).
• b. Two-Way ANOVA – two independent variables.

Output: F-value, p-value. Interpretation: p < 0.05 → significant difference; p > 0.05 → no significant difference.

10. Hypothesis Testing

Definition: Statistical process to determine whether evidence supports a claim.

Types of Hypothesis

• Null Hypothesis (H₀) – no difference exists. Example: Drug A = Drug B
• Alternative Hypothesis (H₁) – difference exists. Example: Drug A ≠ Drug B

Steps in Hypothesis Testing

1. State hypothesis – H₀ and H₁
2. Select significance level – usually α = 0.05
3. Choose statistical test – t-test, ANOVA, Chi-square
4. Calculate test statistic
5. Determine p-value
6. Decision – if p < 0.05 → Reject H₀; p ≥ 0.05 → Accept/Fail to reject H₀

Summary Flow

This structure is commonly used in research methodology, pharmaceutical research, clinical studies, and academic dissertations.

---

— end of document —