True & False Test

All bell-shaped symmetric curves are normal distributions for some mean and standard deviation

false

For the standard deviation of x-bar to be sigma over the square root of n, the population has to be normal

false

If all the values of the data are exactly the same, the mean equals 0

false

t-distributions always have a mean of 0 and a standard deviation of 1

false

the center of a confidence interval is the population parameter

false

A correlation of .75 indicates a relationship 3 times as linear as one with a correlation of .25

false

If the linear model is appropriate, the number of positive residuals will be the same as the numer of negative residuals

false

If the p-value is .015, then the probability that the null hypothesis is true is .015

false

If the sample has variance 0, the variance of the population is also zero. (variance is the standard deviation squared)

false

Increasing the probability of a Type II will increase the power of a hypothesis test, all else being the same

false

Sampling error can be eliminated only if a survey is both extremely well designed and extremely well conducted

false

a 90% confidence interval means that if 100 random samples are taken, 90 will have a confidence interval which indicates the population parameter

false

a correlation of 0.2 means that 20% of the points are highly correlated

false

a larger population size will decrease the length of the confidence interval

false

a smaller sample size will reduce the margin of error in a confidence interval

false

a useful approach to overcome bias in observational studies is to increase the sample size

false

distributions of t-statistics have less spread than the normal. They have less probability in the tails and more in the center than the normal

false

for a single list, outliers are defined starting at the median plus or minus 1.5 times the IQR

false

for the sampling distribution of p-hat to be approx normal, n has to be greater than or equal to 30

false

if a sample size is large enough, the necessarity for it to be a sRS is diminished

false

if bias is present in a sampling procedure, it can be overcome by dramatically increasing the sample size

false

if the p-value is less than the level of significance, then the null hypothesis is proven false

false

if the p-value of a test is .015, the probability that the Ho is true is 1.5%

false

if two events are mutually exclusive, the joint probability is the product of their probabilities

false

perfect correlation, that is, when the points lie exactly on a straight line, results in r

false

sample parameters are used to make inferences about population statistics

false

the law of large numbers says that when n is large enough, the sampling distribution of the sample mean is approximately normal even if the population is not normal.

false

the p-value of a hypothesis test taken when n=N is 0

false

the probability of a type II error will increase as the sample size n increases

false

the sampling distribution of p-hat has a standard deviation of square root of npq.

false

the variance of the set of a sample menas varies directly with the sample size and inversely with the population variance

false

tripling the sample size divides the size of the confidence interval by 3

false

when r=0, there is no relationship between x and y

false

when r=1, there is a perfect cause and effect relationship between the variables

false

when the null hypothesis is rejected. It is because it is not true.

false

you should always examine your data before picking a significance level or deciding on one or two-sided hypothesis test.

false

A distribution spread far to the right side is said to be right skewed

true

Bias has to do with the sampling distribution

true

In all normal distributions, the mean and median are equal

true

Sampling error is the difference between a population parameter and the value of the statistic, related to that parmeter, calculated from a sample

true

The IQR is more resistant to outliers than the range

true

a high confidence level can be obtained no matter the sample size

true

a smaller confidence level will reduce the margin of error

true

by controlling certain variables, blocking can make conclusions more specific in an experiment

true

choosing a significance level alpha sets the probability of a type I errer to exactly alpha

true

choosing a smaller level of significance results in a higher risk of type II error and a lower power

true

for a given population standard deviation, statistics from smaller samples have more variability

true

if all the values of the data are exactly the same, the standard deviation equals 0

true

if the standard deviation of a random variable is zero, it must be true that the random variable takes on only one value

true

in an experiment researchers decide how people are placed into different groups

true

increasing the significance levl will increase the power of a hyptohesis test, all else being the same

true

normal curves with different means are centered around different numbers

true

provided the populaiton is significantly larger than the sample size, the spread of a sampling distribution does not depend on the pop size

true

range of the sample is never greater than the range of the population

true

sampling error is usually smaller when the sample size is larger

true

t-distributions are bell-shaped and symmetric

true

tests of significance (hypothesis tests) are designed to measure the strength of evidence against Ho

true

the area under a normal curve is always equal to one, regardless of what the mean and standard deviation are

true

the area under the t-distribution is 1

true

the area under the z curve between 0 and 2 is half the area between -2 and +2

true

the central limit theorem says that when n is large enough, the sampling distribution of the sample mean is approximately normal even if the population is not normal.

true

the correlation and slope of the regression line have the same sign

true

the greater the degrees of freedom, the closer the t-distribution is to the z-distribution

true

the higher the df, the narrower the tails of the t-distribution

true

the p-value is a conditional probability

true

the p-value of a test is the probability of obtaining a results as extreme or more extreme assuming the null hypothesis is true

true

the sampling distribution of the sample mean is normal if the population is normal.

true

the square of the correlation measures the proportion of the y variability predictable from a linear relationship with x.

true

the standard deviation is never negative

true

the standard deviation of a distribution can never be negative

true

the standard deviation of the sampling distribution of sample means varies directly with the standard deviation of the population and inversely with the square root of sample size

true

while the range is affected by outliers, the IQR is usually not

true