Stats Final T/F review
A dependent variable is characterized by a small number of levels of treatment.
F
A histogram is used to present the frequencies of a qualitative variable.
F
A negatively skewed curve has a thin point directed to the right side of a graph.
F
A parameter is a characteristic of a sample.
F
Bimodal distributions are an example of a normal distribution.
F
Calculating a deviation score requires a standard deviation.
F
Extreme scores on the normal curve are those between the mean and the median.
F
The denominator of a t test is a difference between means.
F
The formula for effect size index, d, has the sample mean in the denominator.
F
The formula for the standard deviation of a sample used to estimate σ has N in the denominator.
F
The mean is an appropriate measure of central tendency for skewed distributions.
F
The mean is an appropriate measure of central tendency with open-ended distributions.
F
The numbers from a nominal scale convey greater concepts of greater than and less than.
F
The researcher chooses the values of the dependent variable in an experiment.
F
To find the overall mean using samples of different sizes, add the two means and divide by 2.
F
Values of d larger than 1.00 are not possible.
F
You can determine the mode of a distribution from a boxplot presented in the textbook.
F
You can determine the standard deviation of a distribution from boxplots presented in the textbook.
F
Your text supports the assertion that you can prove anything with statistics.
F
z scores are used to indicate the central tendency of a distribution.
F
An effect size index of .05 or greater is considered larger.
F (.8 or greater)
If an urn contains 2 red balls, 2 blue balls, and 1 green ball, the probability of drawing a red ball is .50.
F (2/5 = .40)
A univariate distribution is required to calculate a correlation coefficient.
F (bivariate distribution is required)
As N increases, the standard error of the mean increases.
F (decreases - Central Limit Theorem)
The numerator of the z score in Chapter 6 is a standard deviation.
F (deviation score)
Equal distances on the X axis are associated with equal proportions of the normal curve.
F (its a normal curve -bell shaped. Much more area between -1 and 1 than between 3 and 4.)
A bar graph is used to present the frequencies of a quantitative variable.
F (qualitative variable)
By using random samples, uncertainty about the conclusion can be eliminated.
F (reduced but not eliminated)
To know degrees of freedom, you must know the sample mean.
F (usually just N)
A Type 1 error is possible when the null hypothesis is true.
T
A characteristic of the mean is that Σ(x-x̄) =0.
T
A line graph shows the relationship between two variables.
T
A point on the normal curve with .35 of the curve beyond it has .15 between it and the mean.
T
A population with a bimodal distribution could produce a normally distributed sampling distribution.
T
A population with a rectangular distribution could produce a normally distributed sampling distribution.
T
A scatterplot is a graph of a bivariate distribution.
T
A z score of -1.05 is possible.
T
An extraneous variable can influence the values of the dependent variable in an experiment.
T
Bell-shaped and rectangular distributions are symmetrical.
T
Coorelation soefficients range from -1.00 to 1.00.
T
Correlation coefficients cannot be used to establish a cause and effect relationship.
T
Extreme scores on the normal curve are those far from the mean.
T
Finding the interquartile range first requires a calculation of percentiles.
T
For a bivariate distribution with 10 pairs of scores, N+10.
T
For continuous data that are positively skewed, the mean is usually larger than the median.
T
Horizontal axis, x-axis, and abscissa are all names for the same line.
T
If a bivariate distribution is not linear, r will not show the degree of relationship between two variables.
T
If a difference is statistically significant, the null hypothesis was rejected.
T
If the null hypothesis is retained, the difference is not statistically significant.
T
Meaningful confidence intervals of 90, 95, and 99 percent could be calculated for a single sample mean.
T
On a ratio scale, zero means that no amount was present.
T
Sampling distributions are based on random samples.
T
Sampling distributions are the basis of probability statements in statistics.
T
Statements such as "twice as much" and "half as much" are permissible if the variable is measured on a ratio scale.
T
The coefficient of determination is the square of r.
T
The greater the overlap of two distributions, the smaller the size of d.
T
The inflection points on the normal curve are at z=-1 and z=+1.
T
The location of the median is found using the formula, (N+1)/2.
T
The mode is an appropriate measure of central tendency for any of the four scales of measurement.
T
The probability figure that is the conventional cut-off for null hypothesis statistical testing is .05.
T
The reliability of a test can be measured using a correlation coefficient.
T
The total area of each theoretical distribution described in Chapter 6 was 1.00.
T
To determine how much difference there is between two distributions, calculate d.
T
To look up a t value in the t distribution table, you must know sample size.
T
To select a random sample, every member of the population must be identified.
T
To use a sample is to agree to accept some uncertainty about the results.
T
You can determine the 25th and 7th percentile from boxplots presented in the textbook.
T
You can determine the interquartile range of a distribution from a boxplot presented in the textbook.
T
You can determine the skew of a distribution from a boxplot presented in the textbook.
T
z scores are used to describe individual scores.
T
The Central Limit Theorem describes the form of a sampling distribution.
T (CLT - as N increases, sample mean gets closer to μ)
The normal distribution is symmetrical about the median.
T (in a normal distribution, median = mean)
The sample mean is always in the exact middle of a confidence interval.
T (sample mean, not pop. mean)
The regression coefficient, a, tells where the regression line crosses the Y axis.
T (y=a+bx)