Stats Quiz Q/A

Which of the following probability/alpha level orderings refer to most to least probable?

.05, .025, .01, .001

The animal colony in the psychology department contains 20 male rats (15 blue, 5 spotted) and 30 female rats (15 blue and 15 spotted). 50 total rats.....(30 blue, 20 spotted; 20 male, 30 female). What is the probability of selecting a blue male rat at random?

15 out of 50

Z-scores examine one sample mean, T-tests look at ______ sample means(s)

2

If we have a Z score of +1.96, this would be associated with a probability of

2.5 out of 100

The animal colony in the psychology department contains 20 male rats (15 blue, 5 spotted) and 30 female rats (15 blue and 15 spotted). 50 total rats.....(30 blue, 20 spotted; 20 male, 30 female). What is the probability of selecting a blue rat at random?

30 out of 50

The animal colony in the psychology department contains 20 male rats (15 blue, 5 spotted) and 30 female rats (15 blue and 15 spotted). 50 total rats.....(30 blue, 20 spotted; 20 male, 30 female). What is the probability of selecting a female rat at random?

30 out of 50

_______________ statistics describe the data that is on hand, whereas ____________ statistics go beyond the data at hand to make predictions about larger groups of data.

Descriptive; Inferential

A 200% increase in beer consumption compared to a 10% increase in milk consumption for 14-18 year olds indicates that more 14-18 year olds are drinking beer than are drinking milk.

False

A Z score of 0 would tell us that 64% of cases scored lower, while 36% of cases scored higher.

False

A platykurtic distribution has a more peaked center and fatter tails compared to the normal distribution

False

A statistically significant difference is one that has a high probability of occurring, in other words, it is easy to find and likely to occur.

False

A statistician could use data from a small group of beer drinkers to accurately make predictions about heroin users.

False

According to lecture (and most students) Variance values are easier than Standard Deviation values to interpret and explain conceptually.

False

An ordinal scale is an example of a continuous measure.

False

Approximately 95% of cases fall between +1 and -1 standard deviations around the mean.

False

Correlation coefficients based on Z scores are always completely different than those based on Raw Scores.

False

Descriptive Statistics make inferences about larger groups of people from a relatively smaller amount of data.

False

For Z score calculations of correlation, if we use N-1 in the Denominator for the correlation formula, we must have calculated the Z scores themselves by using a Standard Deviation (SD) calculation based on using N, not N-1.

False

If I flip a coin multiple times, the probability of a head turning up goes down exponentially. For example, on the first flip, the odds of a head are 1/2. On the second flip, the odds of a hard are 1/4. On the third flip, the odd of a head are 1/8. And so on.

False

If the population distribution is normal, no matter what, our sampling distribution of sample means will be skewed.

False

If we are using "true" random sampling techniques (i.e., sampling without replacement) in a population of 1000 Democrats, 1000 Republicans, and 1000 Independents, we have to make sure that our sample is collected so that we will have an equal number of Republicans, Democrats, and Independents (e.g., D-R-I-D-R-I).

False

If you don't know the population standard deviation, even if you know the sample standard deviation, you cannot calculate a confidence interval.

False

If you use sampling with replacement, after taking 1 black marble out of a bag of 9 black and 1 white marbles, the odds of taking out another black marble are 1 in 10.

False

In order to be considered a random sample, participants must have an uneven/unequal chance of being selected.

False

Social scientists typically use a population to calculate statistics to answer research questions such as, "who will people likely vote for in the next election?" or "how is depression best treated?"

False

Statistically speaking, extreme scores are very probable or likely to occur.

False

The Mean is the only type of Average.

False

The Null Hypothesis states that there is a relation between two variables or that two means are different.

False

The field of statistics got its start after Psychology became officially recognized as a science in 1935.

False

The mean of a sampling distribution of sample means should be close to, if not equal to, the population mean.

False

The normal distribution is not symmetrical.

False

The null hypothesis for the dependent sample t-test is that the sample mean difference is anything but 0 (i.e., more than or less than 0).

False

The standardized normal distribution/curve is a real curve found in the world.

False

This graph is a fair and accurate representation of comparing how many trucks are still on the road for 3 major automobile manufacturers.

False

We could take SAT scores from College Students and make valid inferences about how well an elderly person would do in a nursing home (and this person never took the SAT or went to college).

False

Ordinal is to Nominal as Ratio is to _________

Interval

"Quantitative" variables include _____________ and ____________ variable types; whereas "Qualitative" variables include ____________ and _____________ variable types.

Interval; Ratio; Nominal; Ordinal

The _____________ is less precise, mathematically speaking than the Median.

Mode

When you have nominal data, you should use the ____________ as the measure of central tendency.

Mode

We're interested in all of the 12th graders in Muncie. Thus, a group of 50 12th graders from all 12th graders in Muncie would be considered a ___________________

Sample

All things being equal, if you RESTRICT, LIMIT, or REDUCE the range of one of the variables (e.g., the Y scores) in a correlation equation, the correlation coefficient (r) becomes smaller.

True

As our sample looks more and more like a population (i.e., the more representative it is), our inferential statistics should be more and more accurate (i.e., the better they will be in saying something about the population).

True

Correlations can be used to determine how reliable a test is (i.e., you can correlate scores at time 1 and time 2 on a test to determine it's reliability).

True

If I had done a pre-test in this course, and everyone took it, then did a post-test in this course, and all the same people took it, I could run a dependent samples t-test on this data to see if there were differences between the pre- and post-test results for the students in the course.

True

If I wanted to analyze differences in opinions about marriage between people, I could conduct either a dependent or an independent t-test depending on how I connected the data (e.g., couples vs male/female differences).

True

If we take a sample and imply something about the population, we are using inferential statistics.

True

OGIVE / Cumulative Distribution Graphs are good methods for estimating, but not calculating, percentiles.

True

Sampling with replacement allows for constant probability in terms of chances of being selected for a sample.

True

Standard deviation squared gives you variance.

True

The null hypothesis essentially states that our findings will not be different from chance.

True

The null hypothesis for ANOVA is that the means you are examining (3 or more) are essentially the same (i.e., equal).

True

The null hypothesis for Dependent Samples T-test is the population mean difference is essentially zero.

True

The null hypothesis for Independent samples T-test is that the two sample means you are examining are essentially the same (i.e., equal).

True

The null hypothesis for a Z-test is that the sample mean and population mean are essentially the same (i.e., equal).

True

The number of questions a professor is asked by students during a Statistics Test could be considered as a variable.

True

The two types of t-tests that we will study are 1) matched/related/dependent and 2) independent.

True

We could expect the percentage of different majors represented in this course (e.g., % of psychology majors) to be about the same as those found in other Psychology 241 (Intro to Stats) courses found on campus at the Ball State University.

True

When using samples, we are depending upon/utilizing inferential statistics to say something about the population.

True

Z tests for determining if a sample is representative of a population are problematic to calculate in the real world because you need to know population information (i.e., population mean and sd) in the formula.

True

A(n) __________ error occurs when we state that there is a difference between groups when there isn't; whereas a(n) __________ error occurs when we state that there is not a difference between groups when there actually is.

Type I ; Type II

Standard deviation is to variance as correlation is to ___________

Variance

If you reject the null hypothesis, you may be saying:

both c and d

In general, as sampling error increases we can:

have less confidence in our inferences to a population

A frequency distribution

is simply a count of how often each value for a particular variable occurs

If a curve or distribution is not "normal", it can display or is said to have ______________ and still be symmetric or ______________ and not be symmetric.

kurtosis; skewness

A platykurtic distribution, compared to the normal distribution,

less peaked at the center with thinner tails

One method of determining the median value for a distribution with an odd number of scores is to:

list all scores from highest to lowest and find the middle most score

The ____________ of a sampling distribution of sample means should equal the _______________ mean.

mean ; population

The 3 measures of central tendency to be discussed in this course are _____, _____, and ______.

mean, median, mode

When thinking about the Standard Error of the Mean and Confidence Intervals, if we do not have the population standard deviation:

none of the above

A(n) ___________ sample is one that is selected so that every unit or case in the population has a(n) _________ chance of being selected

random ; equal

If a student calculated a value for a Z score comparison between 2 groups, and found a value for the Z score to be 3.11, what could that student conclude about the null hypothesis if they were working at the .01 significance level (two tailed)?

reject the null

If a student calculated a value for a Z score comparison between 2 groups, and found a value for the Z score to be 3.11, what could that student conclude about the null hypothesis if they were working at the .05 significance level (two tailed)?

reject the null

In order to randomly select from a population, you must first develop a _______________ .

sampling frame

A(n) _______________ difference is one that has a low probability of occurring, and if such a difference occurs, we would __________ the null hypothesis.

significant ; reject

Statistics can include

the batting average of a baseball player, how effective a particular commercial product is compared to other products, information regarding birth, marriage and/or death frequencies

According to lecture (and most statisticians), "quantitative" U.S. income data (not brackets) would best (and most fairly) be described by using:

the median