Factors that effect the sampling distribution: - correct answer ✔SAMPLE
| | | | | | | | |
DESIGN (SRS): Non random may be biased.
| | | | | | |
SAMPLE SIZE (n): As becomes larger, sample statistic converges on true
| | | | | | | | | |
population.
|
VARIABILTY (SD): How quickly converges. | | | |
Sub-sample populations - correct answer ✔Used to find out about populations.
| | | | | | | | | |
Sample statistics - correct answer ✔Used to predict population parameters.
| | | | | | | | |
SEM (standard error of the mean) - correct answer ✔One of the most
| | | | | | | | | | | |
commonly reported measures of spread. Formula: SEM: SD/sqrt(n)
| | | | | | | |
Xbar - correct answer ✔SAMPLE MEAN. Best average of mean. Xbar is
| | | | | | | | | | |
unbiased. SEM describes the VARIABILITY of Xbar. Regardless of the
| | | | | | | | | |
distribution of individuals in the population, distribution of the sample mean
| | | | | | | | | | |
(Xbar) becomes normal as n gets large. in 95% of all samples, Xbars lies within
| | | | | | | | | | | | | | |
2 SD of the population mean. Less variable & more normal with SRS.
| | | | | | | | | | | | |
Central Limit Theorem - correct answer ✔The theory allows us to infer about
| | | | | | | | | | | |
mean from just 1 sample (using Xbar, s). As n within a random sample becomes
| | | | | | | | | | | | | | |
large, the sampling distribution of a statistic becomes normal, even from a
| | | | | | | | | | | |
strongly non-normal population.
| | |
, Unstandardizing - correct answer ✔Working with inverse normal distribution.
| | | | | | | |
x= zs+Xbar
|
Confidence Intervals - correct answer ✔A formal way to make statements
| | | | | | | | | |
about the probable location of a parameter from a statistic. Larger CI (smaller
| | | | | | | | | | | | |
sample)=less certainty, SMALLER= MORE CERTAINTY. Typically use 90% or
| | | | | | | | |
greater. A general formula to express the confidence interval for the mean:
| | | | | | | | | | | |
(Mean +/+ the standard error). (assuming SD is known):
| | | | | | | | | |
| Xbar(mean)+/- Z*(SD/sqrt(n)) |
What will happen to the interval as % confidence increases? - correct answer
| | | | | | | | | | | |
✔The interval will increase.
| | | |
Standardizing - correct answer ✔All normal curves are the same if we measure
| | | | | | | | | | | |
them in units of size (SD) about the mean. A standardized value is a "z-score".
| | | | | | | | | | | | | | | |
Standard normal curve: mean=0 & SD=1
| | | | |
(POPULATION): (x-mean)/SD |
(SAMPLE): z= (x-xbar)/s | |
Degrees of freedom - correct answer ✔Number of samples minus 1 (n-1).
| | | | | | | | | | |
Null hypothesis (Ho) - correct answer ✔Test designed to assess the strength of
| | | | | | | | | | | |
the evidence against the null hypothesis. A claim about a population
| | | | | | | | | | |
characteristic that is initially asummed to be true. State in terms of "no effect"
| | | | | | | | | | | | | |
or "no difference". Will always be an equality: (u=uo).
| | | | | | | | | |
3 forms: u>uo, u<uo, u(not)=uo.
| | | |
| | | | | | | | |
DESIGN (SRS): Non random may be biased.
| | | | | | |
SAMPLE SIZE (n): As becomes larger, sample statistic converges on true
| | | | | | | | | |
population.
|
VARIABILTY (SD): How quickly converges. | | | |
Sub-sample populations - correct answer ✔Used to find out about populations.
| | | | | | | | | |
Sample statistics - correct answer ✔Used to predict population parameters.
| | | | | | | | |
SEM (standard error of the mean) - correct answer ✔One of the most
| | | | | | | | | | | |
commonly reported measures of spread. Formula: SEM: SD/sqrt(n)
| | | | | | | |
Xbar - correct answer ✔SAMPLE MEAN. Best average of mean. Xbar is
| | | | | | | | | | |
unbiased. SEM describes the VARIABILITY of Xbar. Regardless of the
| | | | | | | | | |
distribution of individuals in the population, distribution of the sample mean
| | | | | | | | | | |
(Xbar) becomes normal as n gets large. in 95% of all samples, Xbars lies within
| | | | | | | | | | | | | | |
2 SD of the population mean. Less variable & more normal with SRS.
| | | | | | | | | | | | |
Central Limit Theorem - correct answer ✔The theory allows us to infer about
| | | | | | | | | | | |
mean from just 1 sample (using Xbar, s). As n within a random sample becomes
| | | | | | | | | | | | | | |
large, the sampling distribution of a statistic becomes normal, even from a
| | | | | | | | | | | |
strongly non-normal population.
| | |
, Unstandardizing - correct answer ✔Working with inverse normal distribution.
| | | | | | | |
x= zs+Xbar
|
Confidence Intervals - correct answer ✔A formal way to make statements
| | | | | | | | | |
about the probable location of a parameter from a statistic. Larger CI (smaller
| | | | | | | | | | | | |
sample)=less certainty, SMALLER= MORE CERTAINTY. Typically use 90% or
| | | | | | | | |
greater. A general formula to express the confidence interval for the mean:
| | | | | | | | | | | |
(Mean +/+ the standard error). (assuming SD is known):
| | | | | | | | | |
| Xbar(mean)+/- Z*(SD/sqrt(n)) |
What will happen to the interval as % confidence increases? - correct answer
| | | | | | | | | | | |
✔The interval will increase.
| | | |
Standardizing - correct answer ✔All normal curves are the same if we measure
| | | | | | | | | | | |
them in units of size (SD) about the mean. A standardized value is a "z-score".
| | | | | | | | | | | | | | | |
Standard normal curve: mean=0 & SD=1
| | | | |
(POPULATION): (x-mean)/SD |
(SAMPLE): z= (x-xbar)/s | |
Degrees of freedom - correct answer ✔Number of samples minus 1 (n-1).
| | | | | | | | | | |
Null hypothesis (Ho) - correct answer ✔Test designed to assess the strength of
| | | | | | | | | | | |
the evidence against the null hypothesis. A claim about a population
| | | | | | | | | | |
characteristic that is initially asummed to be true. State in terms of "no effect"
| | | | | | | | | | | | | |
or "no difference". Will always be an equality: (u=uo).
| | | | | | | | | |
3 forms: u>uo, u<uo, u(not)=uo.
| | | |
