Outline of topics– 223 Midterm

Your best guide is the Powerpoint lectures and the quizzes. However, because it has been requested, I have created this guide to give you an outline of the topics that have been covered.

For this exam, I will provide all formulas along with a z-table.

You should bring a __simple__ calculator. Be sure it
has a square-root key. You may NOT bring
any kind of calculator if it is part of a data-device (eg iphone or other
organizer). Graphing calculators are allowed.

· Quantitative vs Categorical variables

· Charting data: Which charts do you use for categorical data? Which for quantitative (aka nominal) data?

· Charting data: When to use bar vs pie charts.

· Deceptions or misleading information when using pie charts

· Use of histograms.

o Interpretation

o Limtations/Misinterpretations of histograms

o Difference between histogram and bar chart. For example, a histogram should not have spaces between the bars – unless there is 0 data for that particular bin.

o Impact of skewed data on a histogram

· Outliers: Identification.

· Mean vs Median

· The term “resistant” (as it applies to statistics)

· Quartiles:

o IQR

o 5-Number summary

o Boxplot

o 1.5 rule for outliers

· Standard Deviation

o What the concept means

o “Properties” of s

· Density curves

o Concept

o Interpretation

o How the z-score lets you compare “apples and oranges” (e.g. SAT scores and ACT scores)

o Calculations involving areas/percentages/z-scores etc

o Normal distribution: concept & interpretation, properties

· Normal quantile plot

· Scatterplots:

o When are they used?

o Explanatory vs response variables

o Interpretation (eg what does a negative slope tell us? Strength, Outliers)

o Misinterpretations

o Categorical data as the explanatory variable

· Correlation coefficient including properties of r

· Regression

o What is a regression line? Why is it helpful?

o What is the name of the method we have used to determine the best regression line?

o Understand the y = b0 + b1*x formula

o Know
how to find and calculate b_{0} and b_{1}

o Correlation vs regression

o Extrapolation – problems with

·
Coefficient of determination (R^{2}) –
what does it mean?

· Residuals and residual plots

· Effect of outliers and influentials points

· Why is it important to always plot data? (see example in regression lecture)

· Lurking/Confounding variables

· Difference between association and causation

· Anecdotal data

· Difference between Population vs sample

· Counfounding variable

· “Controls” in experimentation

· Placebos

· Biases: review and understand the examples. Review ways to avoid bias

· Double-blind experimentation

· Randomization (not how to do it, but why it is important)

· Stratification