ISA Chapter Summaries PDF

# Chapter 5.1 - **Numerical Data (Quantitative)** = Data in the form of any number, **ex. weight**. - **Categorical Data (Qualitative)** = Data sorted into groups or categories, **ex. eye colour**. - **Continuous Data** = Decimal #’s, **ex. weight**. - **Discrete Data** = Specific values (whole #’s...

# Chapter 5.1 - **Numerical Data (Quantitative)** = Data in the form of any number, **ex. weight**. - **Categorical Data (Qualitative)** = Data sorted into groups or categories, **ex. eye colour**. - **Continuous Data** = Decimal #’s, **ex. weight**. - **Discrete Data** = Specific values (whole #’s), **ex. eye colour**. ## Types of Categorical Data: - **Ordinal Data** = Qualitative data that can be ranked, **ex. poor, fair, good, etc.** - **Nominal Data** = Qualitative data that can’t be ranked, **ex. eye colour.** ## Variability in Data - Variability in data can occur due to errors in measurement or varying conditions in experiments. ## Interpreting Data - Different people can interpret data in different ways. ## Data Collection - Collect data on more than one variable to see if there’s a relationship between them. # Chapter 5.2 - **Population** = All beings studied. - **Sample** = People elected from a population. - **Variability** = Shows how samples differ from each other. - The more similar the samples are to each other, the lower the variability and the more accurate the samples represent the population. - **Sample selection must be random.** ## Sampling Methods - **Simple Random** - Randomly choose a specific # of people. - Examples: Stratified and systematic. - **Systematic** - Put population in an ordered list and choose people @ regular intervals. - **Stratified** - Divide sample into groups with the same proportions as those groups in the population. - Time & cost efficient. - **Cluster** - Divide population into groups and choose a few groups to survey. - **Multistage** - Divide population into a hierarchy and randomly choose people from each group to survey, **ex. Gr1, Gr2, Gr3, Gr4.** - **Convenience** - Choose easily accessible people from population. - **Advantages:** Gives unreliable results because it omits large portions of population. - **Disadvantages:** Cheap to conduct. # Chapter 5.3 ## Types of Experiments: Observational and Experimental: - **Observational** - Look @ situations and make inferences, **ex. Window of classroom closed.** - **Experimental** - Controlled environment, **ex. medicine.** ## Experiment Groups - **Treatment Group** = Participants receive a specifically measured! treatment. - **Control Group** = Participants don’t get the treatment and are compared to the treatment group. ## Advantages of Experimental Studies - In an experimental study, there’s a greater chance of determining the cause of a particular behaviour. ## 3 Things Needed to Occur To Determine Cause: - **Control:** If there’s an effect, researchers must ensure the effect is not caused by bias in any group. - **Randomization,** so each group has similar demographics. - **Replication:** Should have similar demographics. ## Bias - Bias= can occur cause of sampling technique or data collection method. - **Surveys** are less controlled than experiments. - **No opinion** - **Surveys should be anonymous, but can ask for specific demographic info.** - **Surveys should be clear, concise and free of bias.** # Chapter 5.4 - **Primary Source Data:** When you collect and analyze data not summarized or manipulated. - **Microdata:** Individual responses in a survey. - **Secondary Source Data:** Data used by someone other than those who actually collected them. - Usually aggregate data = data that’s combined and summarized so that the microdata can no longer be determined. # Chapter 5.5 ## Bias - **Response Bias:** When respondents change their answers to influence the results, to avoid embarrassment or to give the answer the questioner wants. - **Sampling Bias:** When the sample doesn’t represent the population. **Ex. gym survey**. - **Measurement Bias:** When the collection method is such that the characteristics are consistently over or under represented. - **Leading question:** Or multiple choices are too limiting to give an honest answer. - **Non-response Bias:** Not many people respond and those who do, have very strong opinions on the subject. **Ex. small sample**. ## Different Ways of Displaying Data - Different ways of displaying data can distort it or make it biased.