Chapter 1 | Scientific Investigation

What is hypothesis testing?

A way to check if an idea (hypothesis) about a group of people or things is true, using collected information.

What two main ideas do we test in hypothesis testing?

• Null Hypothesis (H₀): Says there's no effect or difference.

• Alternative Hypothesis (H₁): Says there is an effect or difference.

What are the six steps to test a hypothesis?

1. Clearly state your main assumption (H₀) and what you think might be true instead (H₁);

2. Decide how much risk you're okay with for being wrong (significance level α);

3. Pick the best math test for your data;

4. Figure out the 'danger zone' where your results would be unusual;

5. Collect your data and run the math test;

6. Compare your test result to the danger zone and decide if you should reject H₀ or not.

What is a null hypothesis (H₀)?

The starting assumption, often that nothing special is happening

What is an alternative hypothesis (H₁)?

The opposite of the null hypothesis, suggesting something is happening

What is the significance level (α)?

The chance we're willing to take of making a mistake by saying there's an effect when there isn't one (common choices are 0.05 or 0.01).

What is a Type I error?

This happens when you wrongly decide to reject the null hypothesis, even though it was true.

How do you pick the right test statistic?

It depends on the kind of data you have (e.g., a 't-test' for averages, 'chi-square' for categories).

What is the absolute first step when planning an experiment?

Clearly define the question or problem you want to explore.

What ideas should you form when designing an experiment?

Create a null hypothesis (H₀) and an alternative hypothesis (H₁), which you can actually test with your experiment.

What should be carefully planned when designing an experiment?

How you'll do it: what you'll change (variables), what you'll keep the same (controls), and the exact steps (procedure).

What is crucial when actually doing the experiment?

Follow your plan exactly, avoid unfair influences (bias), and make sure everything is done consistently.

What is data analysis?

The process of organizing and looking closely at information to find meaning and reach conclusions.

Where can we get data from?

Surveys, sensors, and even experiments themselves.

What are the two basic types of data?

Quantitative data (numbers) and qualitative data (descriptions or categories).

What are common ways to collect data?

Using surveys, making observations, and conducting experiments.

What two qualities are important when collecting data?

Reliability (getting consistent results) and validity (measuring what you intend to measure).

What techniques help prepare data for analysis?

Cleaning up the data, adjusting it (normalization), and dealing with any missing pieces.

What is the purpose of exploratory data analysis (EDA)?

Using graphs and basic summaries to spot trends, patterns, and unusual points in your data.

What math methods are used in statistical analysis?

Methods that help us make educated guesses about a larger group, like t-tests, finding correlations, and regression.

What tools help us understand and show data?

Visual tools like various graphs and charts (e.g., bar graphs, line graphs, scatterplots).

What is the goal of making decisions based on data?

To use lessons learned from data to make smart, logical choices and plans.

What should you check for to ensure data is good quality?

Look for unfair influences (bias), extreme values (outliers), repeated entries (duplicates), and any missing information.

Why are data security and privacy important?

To keep private information safe and follow all rules about how personal data should be handled.