# Design and evaluate a behavioural study

An eight-step route from a population and outcome to a comparison, power assumptions and an interpretable interval. The associated guide and design readings supply the experimental-design concepts that the statistics demonstrations do not cover.

1. [Lecture 06 : Objectives, Steps and Basic Definitions](https://nptel.ac.in/courses/111104120)
   Begin by separating the people observed from the population the eventual claim concerns.
   Task: Write a target population and a feasible sample; state one limitation in moving from one to the other.

2. [Lecture 07 : Variables and Types of Data](https://nptel.ac.in/courses/111104120)
   Turn the question into an observation before choosing a statistic.
   Task: Define the primary outcome, units, time window and variable type.

3. [Motivation](https://nptel.ac.in/courses/103106120)
   With an outcome defined, narrow the scientific question to a quantity that a statistical comparison can address.
   Task: State the target difference or proportion in ordinary language and explain what it leaves unanswered.

4. [Statistics for Variance and Proportion](https://nptel.ac.in/courses/103106120)
   Compare possible target quantities instead of assuming every study concerns an average.
   Task: Explain why a mean, variability or proportion is the appropriate target for your particular question.

5. [Hypothesis testing of means](https://nptel.ac.in/courses/103106120)
   Now connect the comparison with the relation between observations.
   Task: Sketch a separate-group and a paired version; read the linked design chapter and identify a carryover issue.

6. [Power of hypothesis test](https://nptel.ac.in/courses/103106120)
   Choose the effect you would want a planned test to detect.
   Task: Name a meaningful alternative and distinguish it from the null hypothesis.

7. [Factors affecting hypothesis test](https://nptel.ac.in/courses/103106120)
   Stress-test the planning inputs rather than reporting one sample-size number.
   Task: Describe how your plan would change under greater variability or a smaller relevant effect.

8. [Confidence interval construction](https://nptel.ac.in/courses/103106120)
   Finish with the estimate you would report and the uncertainty that should accompany it.
   Task: Write a mock results sentence with placeholders for an effect estimate and confidence interval, then one sentence about a remaining design limitation.

Original route: Evidence Press, CC BY 4.0. Linked teaching material retains its own rights. No videos or YouTube API payloads are included.