Term
| What are statistics based on? |
|
Definition
|
|
Term
| What are parameters based on? |
|
Definition
|
|
Term
| What is the sampling distrubution of x bar or p hat? |
|
Definition
| the collection of all possible values of x bar or p hat in all possible samples of the same size from the same population |
|
|
Term
| What kind of data do proportions deal with? |
|
Definition
|
|
Term
| What kind of data does Mean deal with? |
|
Definition
|
|
Term
| Why are x bar and p hat considered unbiased estimators? |
|
Definition
| because the mean (expected value) is equal to mu and p |
|
|
Term
| What is the formula for finding the percent probability/zscore? |
|
Definition
| estimator - mean/standard error |
|
|
Term
| What happens in a normal distribution as the sample size increases? |
|
Definition
| The sample means stay the same the data becomes less spread out (standard deviation gets smaller) |
|
|
Term
| What happens when the sample size increases the original population is not normally distributed? |
|
Definition
| With an increase in sample size, the data becomes more nomally distributed, the sample means stay the same, and the data is elss spread out (standard deviation gets smaller) |
|
|
Term
| We can use the Z-table (sample means) when |
|
Definition
| the original problem says the data is normal (even when there is a small sample size) when the problem says the data is normally distributed and there is a large enough sample size n>30, or when the sample size is greater than or equal to 30, even if the problem doesn't say the original population was normally distributed |
|
|
Term
| Z score for 90% confidence interval |
|
Definition
|
|
Term
| Z score for 95% confidence interval |
|
Definition
|
|
Term
| Z score for 99% confidence interval |
|
Definition
|
|
Term
| What are inferential statistics used for? |
|
Definition
| To make informed conclusions about unknown parameters |
|
|
Term
|
Definition
| t-distributions (because population standard deviation is unknown) |
|
|
Term
| What are characteristics of the t-distribution? |
|
Definition
Probabilities are based on degrees of freedom df=(n-1),
they are bell shaped and symmetric around 0,
has thicker tails and is more spread out than z-distribution
the larger the degree of freedom, the closer t-distribution looks like a zdistribution, when df is > or equal to 30, than z and t are approximately the same |
|
|
Term
| The following will be correct |
|
Definition
We are __% confident that true parameter falls between ___ and ___.
Of __ samples constructed ___% are expected to capture the true parameter.
The probabliity mu is in any (ie)90% interval is 90%. Once created, it is 0 or 1.
x bar is always in the interval created
CI discusses average, never specifics |
|
|
Term
| The confidence interval will increase if the confidence level increases |
|
Definition
| the confidence interval with decrease if the sample size increases |
|
|
Term
| Decreasing the sample size will cause the length of the confidence interval to |
|
Definition
|
|
Term
| decreasing the confidence level will cause the length of the confidence interval to |
|
Definition
|
|
Term
| Decreasing the confidence level and sample size at the same time will cause the length of the confidence interval to: |
|
Definition
| Not enough information given |
|
|
Term
| decreasing the confidence interval while increasing the sample size at the same time will cause the length of the confidence interval to |
|
Definition
|
|
Term
|
Definition
| Not sensitive to moderate departures from normality, but no major outliers will work |
|
|
Term
|
Definition
| Tests an assumption about a parameter |
|
|
Term
| What assumptions need to be met for hypothesis testing for proportions? |
|
Definition
| categorical, SRS, large enough sample size (np>15, n(1-p)>15) |
|
|
Term
|
Definition
|
|
Term
|
Definition
| Whether probablitiy has increased p>po, whether it has decreased p |
|
|
Term
|
Definition
| Z score, measures how far the sample falls from the null hypothesis |
|
|
Term
|
Definition
| summarizes the evidence, tells statisticians how unusual the data is given Ho is true, or the probablitiy of obtaining this value given Ho is true, and is between 0 and 1 |
|
|
Term
| If P-value is less than significance level |
|
Definition
| Ho, null hypothesis can be rejected, support of Ha |
|
|
Term
| If P value is greater than significant level |
|
Definition
| Fail to reject Ho or show statistically significant evidence for Ha |
|
|
Term
| Can only talk about an individual for a population when the problem states that |
|
Definition
| the distribution is normal |
|
|
Term
| Proportions are what kind of data |
|
Definition
|
|
Term
| Means are what kind of data |
|
Definition
|
|
Term
| The population and date distributions for a proportion are |
|
Definition
|
|
Term
| In what distribution do proportions have the possibility to be normal? |
|
Definition
|
|
Term
| When can we add two successes and two failures? |
|
Definition
| While trying to find the confidence interval for proportions |
|
|
Term
| Interpretations of confidence intervals for population proportions |
|
Definition
Will say "Estimates population proportion" "true proportion" or "proportion of all population"
CANNOT SAY: SAMPLE PROPORTION |
|
|
Term
| When finding the confidence interval for population means |
|
Definition
Always think T,
it is ok to use the z score when N is greater than 30,
MAKE A PLOT AND MAKE SURE THERE ARE NO MAJOR OUTLIERS |
|
|
Term
| Interpretations of confidence intervals for population mean |
|
Definition
"Population mean score is between ___ and ____" "mean of all scores for the population" and "true mean"
CANNOT SAY: SAMPLE MEAN and it CANNOT LEAVE OUT "MEAN" or "AVERAGE" all together |
|
|
Term
| Confidence Intervals are statements about |
|
Definition
Parameters, not statistics or individuals
Probability applies before we take data, after we use the word confidence |
|
|
Term
|
Definition
|
|
Term
| Ha: P does not equal #, Ha: p > #, Ha: p < # |
|
Definition
|
|
Term
|
Definition
| (estimator - # Null Hypothesis)/standard error |
|
|
Term
|
Definition
| add together to get p value |
|
|
Term
|
Definition
| Some evidence for Ha and against Ho |
|
|
Term
|
Definition
| Strong evidence for Ha, against Ho |
|
|
Term
|
Definition
| very strong evidence for Ha, against Ho |
|
|
Term
| If P value is less than alpha |
|
Definition
|
|
Term
| If p value is greater than alpha |
|
Definition
|
|
Term
| The probability of observing a test statistic at least as extreme as one found in sample data can be determined by looking at |
|
Definition
|
|
Term
| For any hypothesis test, we always |
|
Definition
| assume the null hypothesis is true |
|
|
Term
| If all other values are held constant, the test statistic will ____ and the p - value will ____ |
|
Definition
| The test statistic will increase and the p-value will decrease |
|
|
Term
| The probability of observing a test statistic at least as extreme as one found in sample data can be determined by looking at |
|
Definition
|
|
Term
| How can we get a narrower confidence interval? |
|
Definition
| increase the sample size and decrease the confidence level |
|
|
Term
| Other things being equal, the margin of error of a confidence interval increases as |
|
Definition
| the population standard deviation increases |
|
|
Term
| Which of the following affects the mean of the sampling distribution of the sample proportion? |
|
Definition
| the population proportion, p |
|
|
