Algebra II Unit 7 Mid Study Guide

Research Study Designs and Methodologies

Mr. Smith conducts research to determine if caffeine improves exam performance by dividing a class into a treatment group receiving two cups of coffee and a control group receiving two cups of decaf. Random assignment is critical to such experiments to ensure group equality. Similarly, a study at the University of Helsinki (Finland) monitored $305$ healthy pregnant women regarding chocolate consumption and infant temperament at $6$ months. This Helsinki study is classified as an observational study because researchers did not manipulate variables or assign treatments; consequently, it shows correlation rather than direct causation between chocolate and "positively reactive" infants. Additionally, when considering an ABC News report on the link between family dinners and grades, utilizing a specific subset such as Naperville students may fail to be representative of the broader population due to sampling bias.

Data Visualization and Distributions

Statistical questions often require visualizing data through histograms. Robert, a sixth grader at Roosevelt Middle School, collected sleep data from $29$ peers to investigate typical rest durations, requiring a histogram with a bin width of $1$ hour. Data points collected included values such as $7$, $8$, $5$, $9$, and $12$ hours. In Kaylie’s class, test scores ranging from $67$ to $93$ were recorded to create a distribution with a bin width of $4$ percent. These visualizations help identify the center, spread, and shape of the data, which is essential for determining if a population follows a normal distribution.

Properties and Applications of Normal Distributions

A normal distribution is characterized by its symmetric, bell-shaped curve where the mean, median, and mode are equal. In a study of elite distance runners, body weights followed a normal distribution with a mean of $63.1\,kg$ and a standard deviation of $4.8\,kg$. Calculations for such distributions involve finding proportions of the population within specific ranges, such as weights between $48.7\,kg$ and $67.9\,kg$, or weights above $70\,kg$. Another application is found with Jada at the Chicagoland Speedway, where she recorded $400$ laps with a mean time of $1.84$ minutes and a standard deviation of $0.07$. In this scenario, evaluating performance involves calculating the proportion of laps falling into specific categories, such as times less than $1.72$ minutes to beat Tyler, or times exceeding $2$ minutes.

Potential Sources of Statistical Bias

Data can be misleading if the sample or the context is flawed. For instance, a principal asking students how long they take to eat breakfast might find an average of $4$ minutes, but this is potentially misleading if the sample is restricted to those who actually eat at school or if the survey method introduces response bias. Likewise, an elementary student finding an average of $1$ child per family by surveying her own class is biased because the sample only includes families that already have at least one child (the student themselves), thus ignoring families with zero children and skewing the "typical" family size.