Final Flashcards
A “powerful” statistical test is a test in which _____.
The probability of rejecting a false null hypothesis is high
This statistical procedure is used when one desires to predict the value of a dependent variable from knowledge of the value of an independent variable.
regression analysis
One can get a general idea of whether two means are significantly different if, on a graph, the values of these
do not overlap.
95 % confidence
This term refers to the prediction of “Y” from a known value of “X” that is beyond the range of the actual
data.
extrapolation
Which is false regarding data that are suitable for parametric tests?
sampled data are measured on an ordinal scale
nothing
nothing
R-squared (r2
) is also known as the _____.
coefficient of determination
The central limit theorem states that _____.
the means of samples from a normally distributed population
The outcomes of statistical tests are usually found in this section of a primary literature paper.
results
A t-distribution with infinite degrees of freedom is identical to this distribution.
normal
An observed frequency distribution of a given type will more closely conform to a theoretical frequency
distribution of the same type under this condition.
increased
. Statistical “error” often refers to the level of confidence that one has regarding how well the statistics of
_____ estimate the statistics of _____.
samples, population
The volume of blood (ml) is measured on this scale.
ratio
Which of the following is a type I error?
rejection of a true null hypothesis
The percentage results of political polls as reported on television usually have a “margin of error”
accompanying the percentages. What is a “margin of error?”
95% confidence interval
In an ANCOVA, the covariate is a _____ variable.
continuous
This calculated value is used in conjunction with a statistical table to determine the probability of a null hypothesis being true.
test statistic
The shape of this theoretical probability distribution is determined by the mean and standard deviation.
normal distribution
The cases of this distribution consist of individual sample means taken from a population.
sample distribution
The strength of the relationship in a correlation analysis is shown by this value.
correlation
The goal of this statistical test is to determine if the means of two separate groups are different. Data are not
normally distributed but the group variances are equal.
Mann-Whitney
How many asterisks indicate a significance level of P<0.001?
3
This general term describes the conclusion about any null hypothesis that has been statistically rejected.
significant
H0: σa2= σb2
is the proper null hypothesis for this statistical test.
Bartlett and Levine
This is the alpha level that most biologists use when testing a null hypothesis.
0.05
This principle states that sample means from a normally distributed population will be normally distributed
regardless of sample size
central limit theorem
This is the result when a true null hypothesis is rejected.
type 1 error
The risk of making a Type 2 error can be reduced by ________.
increasing sample size
If the null hypothesis is A=B, the research hypothesis is ____.
A not = B
The goal of this statistical test is to detect differences between the means of repeated measurements on
individuals in one group. Data are skewed and the group variances are unequal.
Wilcoxon
This is the test statistic for a Mann-Whitney test.
U
Nonparametric tests address either questions of differences or questions of _____.
relationships
This parametric test is considered to be robust.
T-test
This is the name of the tabled value of a test statistic at the specified alpha level.
Critical Value
The goal of this statistical test is to detect differences between two dependent means when the data meet
parametric test assumptions.
Paired t-test
T-tests assume that variances between groups are homogeneous. How would you test this assumption?
Bartlett’s
H0: μa ≤ μb is a suitable null hypothesis for this nonparametric test.
Wilcoxon
This is the numerical relationship between the standard error of the mean and the standard deviation.
SE=SD/(sqt n)
This mathematical theorem predicts that sample means from a non-normally distributed population will have
a normal distribution if the sample size is large enough.
central limit theorem
This frequency distribution is basically a normal distribution whose shape varies with sample size.
t-distribution
The goal of this statistical test is to detect differences among variances of skewed data sets.
levene’s
Which test is least powerful? (a) ANOVA; (b) Pearson’s correlation; (c) independent-samples t-test; (d)
paired t-test; (e) Mann-Whitney test
Mann-Whitney
To determine the effect of two independent variables on a dependent variable, what is the advantage of
doing a single two-way ANOVA as opposed to two separate one-way ANOVAs? (a) a two-way ANOVA is
more robust; (b) a two-way ANOVA calculates the effect of a covariate; (c) a two-way ANOVA is easier to
use on a calculator; (d) a two-way ANOVA assesses possible interaction between the independent variables;
(e) a two-way ANOVA provides a test statistic
(d) a two-way ANOVA assesses possible interaction between the independent variables;
In a standard normal distribution, a z-score of _____ on each side of the mean encloses 95% of the cases.
1.96
In a regression analysis, “Y” is the independent variable and “X” is the dependent variable.
false
The shape of a Poisson distribution is determined by the _____.
mean
This is an important measure of data dispersion.
variance
On a SYSTAT dot graph, these graphics portray variation around the mean.
error bars
This is the square root of the sum-of-squares divided by the sample size.
variance
In an interval scale of measurement, values are neither quantitative nor ranked, and there is no mathematical
or value relationship among them.
false
This is an important measure of data central tendency
median
The basic reason scientific knowledge has advanced so remarkably through the years is because many
dedicated scientists have proved thousands of hypotheses and theories
false
. The temperature of a human body in Celsius should be measured on a ratio scale.
false
The various species contained within a particular genus of birds should be measured on a ranked scale.
false
The most common data transformation used in biology is the logarithmic transformation.
true
This is the probability of obtaining two heads with one flip of two coins.
0.25
Data that are influenced by many small and unrelated random effects are frequently normally distributed.
As a consequence, normally distributed data are widespread and common in nature.
true
The discipline of statistics concerns _____.
using quantitative properties of samples to answer questions
about populations; (b) tallying sports information
This standardized expression permits one to directly compare the relative amount of variation associated
with two or more means of one variable.
coefficient of variation
In SYSTAT, this is the preferred quantitative method for students to determine if data are normally
distributed.
Komolgorov-Smirnov test
In an ANOVA, this is the normal variation expected in individuals that is not a result of being part of a “group.” It results from such things as individual genetic makeup and environmental history.
error variance
These provide a graphical portrayal of variation around the mean.
error bars
In this scale of measurement, values are neither quantitative nor ranked, and there is no mathematical or
value relationship among them.
categorical
The age of a viral particle is measured on this scale.
ratio
In a regression analysis, the regression line is fitted to the data points by this method.
least squares
This is a measure of dispersion.
variance
Who made this statement, “Isn’t that what science is all about…eliminating possibilities?”
Sean Connery
A robust statistical test is a test which _____.
is insensitive to deviations from the assumptions
Which is not an example of statistical inference?
calculating a sample mean
“How heart rate relates to oxygen consumption varies from person to person. Age, weight, sex, body
composition, fitness level, and other factors all play a role. Drawing from population models and their own
research, the companies that manufacture heart rate monitors have developed formulas that couple heart rate
with those different variables and massage it all into an estimate of calorie usage. The onboard calculators
found on treadmills, elliptical trainers and other devices use basically the same approach. Depending on the
machines, however, they typically don’t allow you to enter as much information about yourself as a heart
monitor. The machine might ask for your weight and age, for example, but not your sex or an estimate of
your fitness level. Fewer variables mean a rougher guess.” In statistical terms, what is the meaning of the
last sentence, “Fewer variables mean a rougher guess?”
lower r^2
