AP Statistics Exam Review

Description

Flashcards for AP Exam
Emily Lu
Flashcards by Emily Lu, updated 7 months ago
Emily Lu
Created by Emily Lu 7 months ago
18
0

Resource summary

Question Answer
Categorical Data qualitative data that identifies characteristics (hair color, preferences, etc).
Numerical Data quantitative data; split into discrete (counts) and continuous (measurements).
Uni-variate Data data that describes 1 characteristic of a population.
Bi-variate Data data that describes 2 characteristics of a population.
Multivariate Data data that describes >2 characteristics of a population.
Relative Frequency Formula frequency/n * 100
Interquartile Range (IQR) Q3 - Q1
Describing Categorical Data Identify the highest frequency and lowest frequency,
Name this graph and identify the type of data it's used for Bar Chart, Categorical Data
Name this graph and identify the type of data it's used for Double Bar Chart, Categorical Data w/ 2 or more Groups
Name this graph and identify the type of data it's used for Stacked Bar Chart, Categorical Data (% to whole relationship)
Name this graph and identify the type of data it's used for Pie Chart, Categorical Data (relation of part to whole)
Name this graph and identify the type of data it's used for Dot Plot, Discrete Numerical Data
Name this graph and identify the type of data it's used for Stem-and-Leaf Plot, Univariate Numerical Data
Name this graph and identify the type of data it's used for Split Stem-and Leaf Plot, Univariate Numerical Data w/ long list of leaves
Name this graph and identify the type of data it's used for Back-to-Back Stem-and-Leaf Plot, Univariate Numerical Data w/ 2 Grou[s
Name this graph and identify the type of data it's used for Discrete Histogram, Univariate Discrete Numerical Data
Name this graph and identify the type of data it's used for Continuous Histogram, Univariate Continuous Numerical Data
Name this graph and identify what it's used for Cumulative Relative Frequency Plot, Percentiles
Parameter fixed value about a population.
Statistic calculated value from a sample.
Degrees of Freedom # of observations free to vary; n - 1
Left-Skewed tail on the left.
Right-Skewed tail on the right.
Uni-modal 1 peak.
Bi-modal 2 peaks.
Multimodal >2 peaks.
Linear Transformation Rule +/- constant changes the mean. ×/÷ a constant changes BOTH the mean and SD.
Empirical Rule 68% of values are within 1 SD of the mean. 95% of values are within 2 SD of the mean. 99.7% of values are within 3 SD of the mean.
Combining Means µa+b = µa + µb µa-b = µa - µb
Combining Standard Deviations σa±b = √σ²a + σ²b
Trimmed Mean 1. List values in order. 2. Do % trimmed(n) 3. Remove that many observations from BOTH ends. 4. Calculate mean with the new data set.
Boxplot Outliers any values < Q1 - 1.5(IQR) or > Q3 + 1.5(IQR)
5 Number Summary 1. Minimum 2. Q1 3. Median 4. Q3 5. Maximum
Describing Numerical Data (SUCS) Shape Unusual values Center Spread
Name this graph and identify the type of data it's used for Boxplot, Univariate Numerical Data
Name this graph and identify the type of data it's used for Modified Boxplot, Univariate Numerical Data
Identify the concept that this image displays Empirical Rule
Counting how many ways can an event occur? (order matters/doesn’t)
Permutation order ALWAYS matters.
Combination rder DOESN'T matter.
Union (A ∪ B) the event of A OR B happening
Intersection (A ∩ B) the event of A AND B happening
Disjoint (Mutually Exclusive) 2 events have no outcomes in common.
Independent one event occurs and doesn’t change the probability of another event
Hypothetical 1000 suggests that the overall total will be 1000; used for probability tables only.
Permutation Formula math → prob → nPr
Combination Formula math → prob → nCr
Disjoint Union Formula P(A ∪ B) = P(A) + P(B)
Non-Disjoint Union Formula P(A ∪ B) = P(A) + P(B) - P(A ∩ B)
Independent Intersection Formula P(A ∩ B) = P(A) x P(B)
Not-Independent Intersection Formula P(A ∩ B) = P(A) x P(B|A)
Probability Formula (Equallly Likely Outcomes) favorable outcomes/total outcomes
Conditional Probability P(B|A) = P(A ∩ B)/P(A)
At Least One Formula P(at least one) = 1 - P(Aᶜ)
Exactly One Formula P(exactly one) = P(A ∩ Bᶜ) + P(Aᶜ ∩ B)
Explanatory Variable x-value/independent variable/causes the change.
Response Variable y-value/dependent variable/the outcome of the change.
Correlation relationship between bivariate variables, whether positive/negative
Correlation Coefficient (r) a quantitative assessment of STRENGTH and DIRECTION of a LINEAR relationship.
Least Squares Regression Line (LSRL) the line of best fit, defined by ŷ = a + bx
Extrapolation the LSRL can’t be used for predictions made outside the range (too high/too low).
Coefficient of Determination (r²) the proportion of variation in y determined by the linear relationship between x and y
Residual vertical deviation between a point and the LSRL; y - ŷ
Residual Plot a scatter plot of (x, residual) pairs, determining whether a linear model is appropiate (linear pattern) or not (not a linear pattern).
Influential Point a point that if removed, changes the slope, y-intercept, and/or correlation substantially
High Leverage Point an influential point that changes the slope/y-intercept, affecting the LSRl directly.
Outlier an influential point that changes r.
a (y-int) Formula a = ȳ - bx̄
b (slope) Formula b = r(Sy/Sx)
Interpreting the Correlation Coefficient (r) "There is a (weak/moderate/strong) (negative/positive) linear relationship between x and y."
Interpreting the Slope "For a one unit increase in x, there is a predicted (increase/decrease) of b in y."
Interpreting the Coefficient of Determination (r²) "Approximately r²% of the variation in y is explained by the LSRL of x on y."
Identify the type of plot based on this image Scatter Plot
Census a complete count of the population.
Sampling Design method used to choose a sample from the population.
5 Types of Sampling Design 1. Simple Random Sample (SRS) 2. Stratified Random Sample 3. Systematic Random Sample 4. Cluster Sample 5. Multistage Sample
Sampling Frame a list of every individual in the population.
Simple Random Sample each individual/set of individuals has an EQUAL chance of being selected
Stratified Random Sample population is divided into STATA. An SRS is taken from each strata.
Systematic Random Sample randomly selects a BEGINNING POINT and follows a systematic approach.
Cluster Sample randomly picks a location and samples ALL from there.
Multistage Sample splits the process into stages and takes an SRS at each stage.
Bias a systematic error in measuring the estimate; often favors certain outcomes.
6 Types of Bias 1. Voluntary Response 2. Convenience Sampling 3. Undercoverage 4. Nonresponse 5. Response Bias 6. Wording of Questions
Voluntary Response SELF-SELECTION; people choose to respond because they have strong opinions.
Convenience Sampling asking the easiest people to participate.
Undercoverage when certain groups from the population are left out of the selection process.
Nonresponse when an individual chosen refuses to participate/can’t be contacted.
Response Bias when the respondent/interviewer causes bias by giving the wrong answer.
Wording of Questions the use of big words/connotation can cause bias through confusion and indirect persuasion.
Observational Study observing outcomes WITHOUT treatment.
Experiment observing outcomes AFTER treatment.
Survey simply asking respondents for data; NO observations or treatment.
Experimental Unit the individual to which the different treatments are assigned.
Factor x; what are we testing?
Response Variable y; what are we measuring?
Level a specific value for the factor that splits it into different categories.
Treatment a specific experimental condition applied to the units.
Control Group a group used to compare the factor against.
Placebo dummy treatment with no effect.
Blinding experimental units don’t know which treatment they’re getting.
Double Blind neither the experimental units nor the evaluator know which treatment was used.
Confounding Variable outside variable that affects the outcome but wasn’t considered in the beginning.
Block homogeneous group formed by experimental units that share similar characteristics.
3 Types of Experimental Design 1. Completely Randomized 2. Randomized Block 3. Matched Pairs
Completely Randomized experimental units are assigned randomly to treatments.
Randomized Block experimental units are blocked into homogeneous groups. Then, they are randomly assigned to treatments.
Matched Pairs units are paired up; one gets treatment A and the other automatically gets treatment B. OR, every experimental unit gets both treatments in a random order.
5 Parts of a Simulation 1. Model 2. Trial 3. Assumptions 4. Chart 5. Conclusion
Model Random Digit Table “Let (digits) represent _______. “
Trial “I will select (# of single digit/double digit numbers) to represent (each unit/group). I will record ____ and perform 5 trials.”
Assumptions "P(probability) = #" List all probabilities.
Chart Draw a chart displaying the trials and what you're testing. Then, sum up your results and divide it by the number of trials to achieve your approximately results.
Conclusion “Based on my simulation, I estimate… (approximate results).”
Binomial Distribution tests for the number of successes that can occur out of a given number of trials.
Geometric Distribution tests for the number of trials until the 1st success is reached.
pdf used when looking for exact values; P(X = x)
cdf used when looking for cumulative values; P(X </≤/>/≥ x)
Mean of Linear Function μᵧ = a + b(μₓ)
Standard Deviation of Linear Function σᵧ = |b|σₓ
Unusual Distribution a continuous distribution with uniquely shaped density curve composed of triangles, rectangles, and/or trapezoids.
Uniform Distribution an evenly distributed continuous distribution; shaped as a rectangle.
Normal Distribution a continuous distribution with a symmetrical bell-shaped density curve defined by the mean and standard deviation.
Standard Normal Distribution a normal distribution with mean of 0 and standard deviation of 1.
Normal Probability Plot a scatter plot used to assess normality; linear pattern = distribution is approximately normal.
Trapezoid Formula A = 1/2(b1+b2)h
Rectangle Formula A = bh
Triangle Formula A = 1/2bh
Height of Uniform Dist. 1/(b - a)
Probability of Uniform Distribution A = bh
Mean of Uniform Dist. μₓ = (a+b)/2
Standard Deviation of Uniform Dist. σₓ = √((b-a)²/12)
Probability of Normal Dist. normcdf(l, u, μ, σ)
X-value of Normal Dist. invNorm(a, μ, σ)
Standardization Formula z = (x -μ)/σ
When SD increases, what happens to the normal curve? It flattens and spreads out.
When SD decreases, what happens to the normal curve? It becomes narrower and thinner.
Identify the type of density curve based on this image Unusual Density Curve
Identify the type of density curve based on this image Uniform Density Curve
Identify the type of density curve based on this image Normal Density Curve
Identify the type of density curve based on this image Standard Normal Density Curve
Sampling Variability the observed value of the statistic depends on the particular sample selected from the population.
Point Estimate statistic used to estimate a parameter; often not close to the true value of the parameter.
Confidence Interval interval of possible values for the population characteristic.
Confidence Level the success rate of all confidence intervals that contain the true proportion p.
Confidence Interval Default Formula point estimate ± critical value(standard error)
Z-score Formula 1. (1 - CL)/2 = a 2. 2nd → vars → invNorm(a, 0, 1, left) → Use the POSITIVE vers.
number of successes/n
Null Hypothesis (H0) a claim about the parameter initially assumed to be true.
Alternate Hypothesis (Ha) competing claim against the null; what you are trying to prove.
Test Statistic indicates how many standard deviations the statistic is from the parameter.
P-value the probability of obtaining a test statistic as inconsistent as the null hypothesis, assuming it’s true.
Level of Significance (α) he probability that we REJECT the null hypothesis, assuming it’s true.
Test Statistic Default Formula (statistic - parameter)/standard error
P-value for Proportions 2nd → vars → normcdf(l, u, 0, 1)
np Tests If np ≥ 10 and n(1-p) ≥ 10), the sampling distribution is approximately normal.
Type I Error rejecting the null hypothesis when it’s true; denoted by alpha (α).
Type II Error failing to reject the null hypothesis when it’s false; denoted by beta (β).
Consequences the outcomes of making Type I/Type II errors.
Relationship between α and β α and β are inversely related; as α gets bigger, β gets smaller, vise versa.
Power the probability that the test rejects the null hypothesis when the alternate hypothesis is true (CORRECT).
What happens if alpha increases? Power increases, Type I Error increases, and Type II decreases.
What happens if n increases? Power increases and Type II Error decreases.
What happens if P0 - Pa increases? Power increases and Type II Error decreases.
np Tests (2-Prop) If n₁p₁ ≥ 10, n₁(1-p₁) ≥ 10), n₂p₂ ≥ 10, and n₂(1 - p₂) ≥ 10, the sampling distribution is approximately normal.
Central Limit Theorem when n ≥ 30, the sampling distribution can be approximated by a normal curve.
t Distribution a continuous distribution based on df; used when σ s unknown.
df (t-Dist) df = n - 1
P-value for Means 2nd → vars → tcdf(l, u, df)
T-score Formula 1. (1 - CL)/2 = a 2. 2nd → vars → invT(a, df) → Use the POSITIVE vers.
Central Limit Theorem (2-Samp) when n₁ ≥ 30 and n₂ ≥ 30, the sampling distribution can be approximated by a normal curve.
Pooled t Inference used when the variances of 2 populations are equal; σ₁ = σ₂
df (Matched Pairs) df = n - 1
df (2-Samp) Use 2-SampTTest and truncate the value.
k the number of categories.
χ² test tests the counts of CATEGORICAL data; the 3 types are GOF, homogeneity, and independence
GOF test measures univariate data for a single sample; uses a ONE-WAY table.
Homogeneity measures univariate data for TWO/MORE SAMPLES; uses a two-way/more table.
Independence measures BIVARIATE data for two/more samples; uses a two-way/more table.
Expected Counts (GOF) n(proportion)
Expected Counts (Homoegeneity and Independence) Make a matrix and use χ²-Test
df (GOF) df = k - 1
df (Homogeneity and Independence) df = (r- 1)(c - 1)
P-value (Chi-squared) 2nd → vars → χ²cdf(χ2, ∞, df)
Identify the type of distribution based on this image χ² Distribution
Deterministic Relationship a relationship in which the value of y is determined by the value of x.
Error Variable (e) a random deviation that causes observed (x, y) points to avoid falling exactly on the population regression line.
Test Statistic (LinReg) t = b/sb
P-Value (LinReg) 2nd → Vars → tcdf(l, u, df)
df (LinReg) df = n - 2
Show full summary Hide full summary

Similar

CUMULATIVE FREQUENCY DIAGRAMS
Elliot O'Leary
TYPES OF DATA
Elliot O'Leary
Pythagorean Theorem Quiz
Selam H
Statistics Key Words
Culan O'Meara
Geometry Vocabulary
patticlj
SAMPLING
Elliot O'Leary
Algebra 2 Quality Core
Niat Habtemariam
FREQUENCY TABLES: MODE, MEDIAN AND MEAN
Elliot O'Leary
HISTOGRAMS
Elliot O'Leary
GROUPED DATA FREQUENCY TABLES: MODAL CLASS AND ESTIMATE OF MEAN
Elliot O'Leary