Mean and Median
Requires a browser that supports Java 1.1
This applet demonstrates basic properties of the
mean and median including (a) the effect of skew on the relative size of the
mean and median, (b) the mean deviation from the mean is zero, and (c) the mean
squared deviation from the mean is less than or equal to the mean squared deviation
from the median (or any other number).
Concepts: central tendency, mean, median, skew, least squares.
Sampling
Distribution Simulation
Requires a browser that supports Java
This applet estimates and plots the sampling distribution of various statistics.
You specify the population distribution, sample size, and statistic. An animated
sample from the population is shown and the statistic is plotted. This can be
repeated to estimate the sampling distribution.
Concepts: sampling distribution, standard
deviation, standard error, central limit theorem, mean, median, efficiency,
fluctuation, skew, normal distribution.
Confidence
Intervals
Requires a browser that supports Java 1.1.
Confidence intervals on the mean are generated for simulated experiments. The
confidence level and sample size can be manipulated.
Concepts: confidence interval, mean, standard deviation.
Normal
Approximation to the Binomial Distribution
Requires a browser that supports Java 1.1
This demonstration allows you to view the binomial distribution and the normal
approximation to it as a function of the probability of a success on a given
trial and the number of trials. It can be used to compute binomial probabilities
and normal approximations of those probabilities.
Concepts:
binomial distribution, normal distribution, central limit theorem, correction
for continuity.
Confidence
Interval on a Proportion
Requires a browser that supports Java 1.1.
The effect of violating the assumption that the sampling distribution of p is
normal can be investigated by varying N and Pi.
Concepts:
binomial distribution, normal distribution, central limit theorem, confidence
interval.
Components of
r
Requires a browser that supports Java.
The slope, standard error of the estimate, and the standard deviation of X can
all be manipulated independently to see the effect on the scatterplot and on
r.
Concepts: Correlation, slope, standard error of the estimate, variance,
restriction of range, proportion of variance explained.
Regression
by Eye
Requires a browser that supports Java 1.1
A scatterplot is displayed and you draw in a regression line by hand. You can
then compare your line to the best least squares fit. You can also try to guess
the value of Pearson's correlation coefficient.
Concepts: Correlation, regression line, mean squared error.
Restriction
of Range
Requires a browser that supports Java 1.1
The range of X can be manipulated to investigate its effect on Pearson's r and
other aspects of the relationship between X and Y.
Concepts: Correlation, restriction of range, slope, standard error of estimate.
Repeated
Measures
Requires a browser that supports Java 1.1
This applet lets you investigate differences between correlated and independent
t tests.
Concepts: t test, within-subject
variable, between-subject variable, power.
A "Small" Effect Size
Can Make a Large Difference
Requires a browser that supports Java.
This applet demonstrates that even a "small" effect can be important
under some circumstances. Applicants from two groups apply for a job. The user
manipulates the difference between groups on the variable on which selection
is made and the cutoff for hiring. The effects on the proportion of hired applicants
from each group are displayed. A related phenomenon is discussed in the article:
Martel, R., Lane, D. M., & Willis, C. (1996) Male-female differences: A
computer simulation.
American Psychologist, 51, 157-158.
Concepts: normal distribution,
selection ratio, effect size, omega squared, proportion of variance explained.
Chi Square Test of Deviations
from Expected Frequencies
Requires a browser that supports Java
1.1
You can specify whether you wish to sample from a uniform or a normal distribution.
The applet does the sampling and tests the significance of deviations from these
two distributions.
Concepts: Goodness of fit,
Chi Square, normal distribution, uniform distribution.
2 x 2 Contingency Tables
Requires a browser that supports Java
1.1
Simulates experiments using 2 x 2 contingency tables. You specify the population
proportions and the sample size and examine the effects on the probability of
rejecting the null hypothesis.
Concepts: Chi Square, correction
for continuity, normal approximation..
Normal Approximation to the Binomial
Distribution (Old Version)
Requires a browser that supports Java.
Histograms, Bin Widths, and Cross Validation
Requires a browser that supports Java 1.1.
Demonstrates how a histogram is affected by bin width and starting point of first bin. Illustrates cross-validation criterion for assessing histograms.
concepts: histogram, bin width, cross validation, density estimation.