Taxonomic Skills

Unit 4 : Sampling Techniques and Experimental Design

Unit 1: Categorical Data

CALCULATE While Using 2-Way Tables:

Find Joint Relative Frequency, Joint Conditional Relative Frequency, Conditional Relative Frequency, Marginal Relative Frequency; and Analyze for Independence

DESCRIBE the Association from a:

Ribbon Segmented Bar Graph: Side-by-Side Bar Graph

DRAW:

Side-by-Side Ribbon Segmented Bar Graph with Perfect Independence; Side-by-Side Bar Graph

Unit 2: Univariate Quantitative Data

Describe:

Distribution from a histogram, stemplot,

Compare & Contrast:

Displays; Grouped boxplots; Boxplots; Timeseries; Histograms; z-scores; Effectiveness of Boxplots vs. Stemplots vs. Dotplots; Dotplots; Back-to-back stemplots; Appropriateness of mean vs. median

Calculate:

Sum of median of grouped boxplots; The proportion of datum on a stemplot; Moving average; z-scores; Sum of z-scores; Mean/Median Ratio; Percentage of datum between 2 x-values; Top or bottom x% of data; Midrange; Measures of spread

Explain:

How changing one value will affect measures of center; How timeseries show variability and moving averages show trend; The effect of linear transformations on boxplots and summary statistics; The steps for finding an outlier

Draw:

A moving average on a timeseries; A stemplot; Side-by-Side Dotplots; Boxplot from split data/summary statistics; Parallel Boxplots

Identify:

Outliers from a Summary Table; 5-Number Summary on Boxplot; IQR on Boxplot

Label:

5-Number Summary on Boxplot; IQR on Boxplot

Justify:

Based on spread of grouped boxplots; A choice based on center, variability or spread;

Interpret:

Mean/Median Ratio; (Upper 50%)/(Lower 50%) Ratio; Percentile; Flat segments of a cumulative frequency polygon; z-score for likeliness; Values as typical or not

Unit 3: Bivariate Quantitative Data

Describe:

The difference between positive and negative trend; Scatterplots for association; Subsets; Simpson's Paradox;

Compare & Contrast:

Medians of Subsets; Residuals; Residual plots for predictive power; Scatterplots for associations; Pros and Cons of scatterplots vs. histograms; Appropriateness of transformed models

Calculate:

Actual y from x-value and residual; Slope from a table of values; Frequency of points above, on, or below the LSRL; Predicted y from LSRL model; Residual from a Scatterplot and Output; Difference between 2 x-values using LSRL for prediction; Find predicted y from varying reexpressed models (x,lin(y)), (x,sqrt(y))

Explain:

The definition of a bivariate outlier; The definition of linear; Strong vs. weak; Underpredictions vs. overpredictions

Draw:

LSRL on a scatterplot with a broken axis; Scatterplot; Scatterplot showing subsets; From a scatterplot, draw a histogram

Identify:

An Outlier on a Scatterplot; Corresponding values on 2 different scatterplot; Max/Min x and y value on a scatterplot; LSRL model from regression output; Standard Error of Slope; A better transformed model

Label:

Outlier on a scatterplot; Influential point on a scatterplot

Justify:

The appropriateness of inappropriateness of a LSRL for a dataset; Use of interpolation from the model as it pertains to the domain of data; Non-use of extrapolation from the model as it pertains to the domain of data;

Interpret:

y-intercepts; Slope; Coefficient of Determination; Se;

Unit 4 : Sampling Techniques and Experimental Design

Identify and Possibly Suggest Improvements of:

Treatments, Experimental units, Response Variable, Control, Replication, Blinding, Double-Blinding

Compare and Contrast:

Experiments with Observational Studies as a Difference of Treatments, Randomized and Controlled Experiments Suggesting Cause-and-Effect, the Generalization of Results as Being the Result of How Subjects/Experimental Units are Sampled

Select and Explain:

Appropriate Blocks in Context or From a Diagram to Mathematically Reduce Variability Caused by Potential Confounding Variables

Recognize and Recommend Improvements for:

The Effects of the Following Types of Bias: Voluntary Response Bias, Nonresponse Bias, Undercoverage, Self-Selection, Convenience, Response

Identify and Explain:

Biased and Unbiased Estimators as Having Low Spread Centered at a Population Parameter

Illustrate and Explain:

Completely Randomized Design, Randomized Block Design, Matched Pairs Design, and Cross-Over Design

Describe:

A Sampling Procedure Assigning Labels, Using a Random Number Generator, While Considering Whether to Replace or Not Replace Subjects Back into the Sampling Pool

Identify, Explain, and Compare, in Context:

Simple Random Sampling (SRS), Stratification, Cluster Sampling, Cluster and Stratification Sampling with a Diagram, Cluster Sampling vs. Stratification, Simple Random Sampling vs. Stratification

IN PROGRESS

Page updated

Google Sites

Report abuse