Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
36 Cards in this Set
- Front
- Back
Margin Of Error Of A Sample Proportion
|
The margin of error is the half-width of a confidence interbal. The confidence interval for a proportion is: p ̂±z^* √((p ̂(1-p ̂ ))/n)
|
|
Margin Of Error Of A Sample Mean
|
The half-width of a confidence interval for a sample mean. x ̅±t^* s/√n
|
|
Binomial Distribution Formula
|
P(x=k)=(n/k) p^k (1-p)^(n-k)
|
|
Binomial Distribution Expected Value
|
Expected Value = sample size x proportion. EV=np
|
|
Binomial Distribution Standard Deviation
|
σ_x=√(np(1-p))
|
|
Technical Conditions For A 1-prop Ztest OR Interval
|
.The data must be a simple random sample from the population of interest.
.nπ_0≥10 and n(1-π_0)≥10 |
|
Technical Conditions for A T-test OR Interval
|
.The data must be a simple random sample from the population of interest.
.The sample size must be greater than or equal to 30 or it must be normally distributed. |
|
Technical Conditions For A 2-prop Ztest OR Interval
|
.Test:
-The data must be randomly assigned to treatment groups or a simple random sample from the population of interest. -n_1 (p_c ) ̂≥5 AND n_1 (1-(p_c ) ̂)≥5 -n_2 (p_c ) ̂≥5 AND n_2 (1-(p_c ) ̂)≥5 -p ̂_c=(p ̂_1+p ̂_2)/(n_1+n_2 ) .Interval: -The data must be randomly assigned to treatment groups or a simple random sample from the population of interest. -There must be 5 successes and 5 failures in each group. |
|
Technical Conditions For A 2-sample T-test OR Interval
|
.The data must be randomly assigned to treatment groups or a simple random sample from the population of interest.
.Both sample sizes must be greater than or equal to 30 or normally distributed. |
|
Technical Conditions For The Difference Between Two Means (paired) T-test
|
The technical conditions are the same as a 1-sample t-procedure but the observational units are the pairs and the data are the differences between the two groups.
.The data must be a simple random sample from the population of interest. .The sample size must be greater than or equal to 30 or it must be normally distributed. |
|
Technical Conditions For A x^2 GOF Test
|
.The data must be a simple random sample from the population of interest
.The expected counts in each box must be 5. .df=(r-1)(c-1) |
|
Technical Conditions For A x^2 Test Of Independence
|
.The observations are a simple random sample from the population of interest.
.The expected counts are at least 5 in each category. .df=(r-1)(c-1) |
|
Technical Conditions For A x^2 Test Of Equal Proportions Or Homogeneity
|
.The data must be randomly assigned to treatment groups or be a simple random sample from the population of interest.
.The expected counts are at least 5 in each category. |
|
Technical Conditions For The Slope Of A Regression Line OR Interval
|
.The data must be a simple random sample from the population or interest or be randomly assigned to treatment groups.
.The 2 variables must be linearly related. .For any x-value the y-values must be evenly distributed across the x-axis. .The standard deviation of the residuals are evenly distributed b±t^* SE(b) .df=n-2 |
|
Technical Conditions For A Correlation Coefficient Test
|
.The data must be a simple random sample from the population of interest.
.Both variables must be normally distributed t= (r√(n-2))/√(1-r^2 ) |
|
Expected Value Formula And Definition
|
Multiply each outcome by its probability and then add these values over all possible outcomes.
E(x)=∑x_i (p(x_i ) ) |
|
Regression Equation And Meaning
|
The line that achieves the exact minimum value of the sum of squared residuals.
ŷ=a+bx |
|
Formula And Meaning Of Slope
|
Predicted change in response (y) variable associated with a 1 unit increase in explanatory variable.
b=r(s_y/s_x) |
|
Formula And Meaning Of y Intercept
|
When the explanatory (x) variable is 0 this is where the response (y) variable would be.
a=y̅-bx̅ |
|
Outlier Calculation
|
x>Q3+(1.5xIQR)=outlier
x<Q1-(1.5xIQR)=outlier |
|
CLT For The Sampling Distribution Of A Sample Proportion
|
√(p(1-p)/n)
|
|
CLT For The Sampling Distribution Of A Sample Mean
|
σ_x=σ/√n
|
|
Z-score Or Standardization
|
Subtract the mean from the value of interest and divide by the standard deviation to find the z-score.
|
|
General Meaning Of A Confidence Level
|
This is a measurement of how confident you are that the interval does contain the true parameter value. In the long run after repeated sampling and construction of intervals, _% of those intervals will contain the true population parameter.
|
|
General Meaning Of A Confidence Interval With Magnitude
|
Used to estimate the true value of a population parameter.
|
|
General Meaning Of Probability
|
Proportion of times an event would occur if the random process was repeated many times.
|
|
General Meaning Of Statistical Significance
|
A sample result is said to be statistically significant if it is unlikely to occur due to random sampling variability alone.
|
|
Forms Of Bias
|
Sampling Bias:
.sampling bias .voluntary response bias .convenient sampling bias .judgemental bias Survey Bias .response bias .non-response bias .wording bias .undercoverage |
|
Probability Addition Rule
|
Pr(E or F)=Pr(E) + Pr(F) - Pr(E and F)
|
|
Probability Multiplication Rule
|
Pr(A and B)=Pr(A) x Pr(B)
|
|
Conditional Probability
|
This is the chance that one event will occur given that the second event has already occured. Pr(A/B)=Pr(A and B)/Pr(B)
|
|
Independence In Probability
|
This is the likeliness that one event is unaffected by another.
Pr(A and B)=Pr(A)xPr(B) |
|
Mutually Exclusive In Probability
|
Two events are mutually exclusive if it is impossible for them to occur together. Pr(A or B)=Pr(A)+Pr(B)
|
|
Relative Risk Meaning And Interpretation
|
p̂_A/p̂_B =relative risk
This is how much more effective one thing is than another. The risk you are taking with one treatment over another. |
|
Type 1 Error
|
This is when you reject the null hypothesis when it is actually true. "False Alarm"
|
|
Type 2 Error
|
This is when you fail to reject the null hypothesis when it is actually false. "Missed Opportunity"
|