P-Value & Data Collection – Quick Review

P-Value

• Probability of observing the sample (or more extreme) when $H_0$ is true.
• Use two-sided test unless opposite direction is irrelevant.
• Decision rule: if P < \alpha (e.g., $0.05$) → reject $H_0$; otherwise retain.
• Small $P$ indicates statistical evidence, not practical importance.

Significance, Errors & Power

• Significance level $\alpha = P(\text{Type I error})$ (false positive).
• Type I error: reject true $H_0$.
• Type II error: accept false $H_0$; probability $\beta$.
• Power $=1-\beta$; boosted by larger sample & appropriate test.
• Decision matrix:
  • $H<em>0$ true → correct $1-\alpha$ / Type I $\alpha$. • $H</em>0$ false → Type II $\beta$ / correct (power).

Data Collection Concept

• Systematic gathering of information to answer research questions & test hypotheses.
• Requires clear variables, sampling, suitable instruments, ethics & accuracy.

Data Types

Qualitative

• Descriptive, non-numerical; explores “how/why”.
• Methods: focus groups, interviews, observation, document review.
• Rich insight, context-specific, costly, not generalizable.

Quantitative

• Numerical; measures “what”. Scales: nominal, ordinal, interval, ratio.
• Methods: surveys (closed questions), experiments, records.
• Cheaper, comparable, effect size measurable; limited explanatory depth.

Data Classification

• Primary Data: first-hand, original; higher validity.
• Secondary Data: pre-existing; quicker but potential bias/obsolescence.

Primary Data

• Main sources: experiments, surveys, questionnaires, interviews, observations.
• Advantages: tailored, controlled quality, possibility of extra data.
• Disadvantages: planning, cost, ethics, data-collection burden.

Secondary Data

• Sources: books, censuses, organizational records, archives, journals, databases.
• Advantages: inexpensive, fast, established validity, baseline for comparison.
• Disadvantages: relevance, accuracy, dated, authenticity & copyright issues.

Key Primary Data Collection Methods

• Questionnaires
• Structured/semistructured interviews
• Focus group discussions
• Direct observation
• Surveys (field/online/telephone)
• Case studies & diaries
• Time–motion/activity sampling
• Experimental & statistical techniques