FORMULA CARD FOR WEISS’S ELEMENTARY STATISTICS, FOURTH EDITION Larry R. Griffey • Linear correlation coefficient: 1 6(x − x)(y − y) r H n−1 sx sy NOTATION In the formulas below, unless stated otherwise, we employ the following notation which may or may not appear with subscripts: σ d p̂ p O E n H sample size x H sample mean s H sample stdev Qj H j th quartile N H population size µ H population mean H population stdev H paired difference H sample proportion H population proportion H observed frequency H expected frequency Sxy rH p Sxx Syy CHAPTER 5 Probability and Random Variables • Probability for equally likely outcomes: f , P (E) H N where f denotes the number of ways event E can occur and N denotes the total number of outcomes possible. CHAPTER 3 Descriptive Measures • Special addition rule: 6x • Sample mean: x H n P (A or B or C or · · · ) H P (A) + P (B) + P (C) + · · · (A, B, C, . . . mutually exclusive) • Range: Range H Max − Min • Complementation rule: P (E) H 1 − P (not E) • Sample standard deviation: s 6(x − x)2 or sH n−1 • Quartile positions: (n + 1)/4, s sH 6x 2 − (6x)2 /n n−1 (n + 1)/2, 3(n + 1)/4 • General addition rule: P (A or B) H P (A) + P (B) − P (A & B) • Mean of a discrete random variable X: µ H 6xP (X H x) • Standard deviation of a discrete random variable X: p p σ H 6(x − µ)2 P (X H x) or σ H 6x 2 P (X H x) − µ2 • Interquartile range: IQR H Q3 − Q1 • Lower limit H Q1 − 1.5 · IQR, • Factorial: k! H k(k − 1) · · · 2 · 1 n n! • Binomial coefficient: H x! (n − x)! x Upper limit H Q3 + 1.5 · IQR • Population mean (mean of a variable): µ H 6x N • Binomial probability formula: • Population standard deviation (standard deviation of a variable): s s 6(x − µ)2 6x 2 σ H or σ H − µ2 N N • Standardized variable: z H or n x P (X H x) H p (1 − p)n−x , x where n denotes the number of trials and p denotes the success probability. x−µ σ • Mean of a binomial random variable: µ H np CHAPTER 4 Descriptive Methods in Regression and Correlation • Standard deviation p of a binomial random variable: σ H np(1 − p) • Sxx , Sxy , and Syy : Sxx H 6(x − x)2 H 6x 2 − (6x)2 /n CHAPTER 7 The Sampling Distribution of the Mean Sxy H 6(x − x)(y − y) H 6xy − (6x)(6y)/n • Mean of the variable x: µx H µ Syy H 6(y − y)2 H 6y 2 − (6y)2 /n √ • Standard deviation of the variable x: σx H σ/ n • Regression equation: ŷ H b0 + b1 x, where Sxy 1 b1 H and b0 H (6y − b1 6x) H y − b1 x Sxx n • Standardized version of the variable x: x−µ zH √ σ/ n • Total sum of squares: SST H 6(y − y)2 H Syy CHAPTER 8 Confidence Intervals for One Population Mean 2 • Regression sum of squares: SSR H 6(ŷ − y)2 H Sxy /Sxx • z-interval for µ (σ known, normal population or large sample): σ x ± zα/2 · √ n 2 • Error sum of squares: SSE H 6(y − ŷ)2 H Syy − Sxy /Sxx • Regression identity: SST H SSR + SSE • Coefficient of determination: r 2 H σ • Margin of error for the estimate of µ: E H zα/2 · √ n SSR SST -1- FORMULA CARD FOR WEISS’S ELEMENTARY STATISTICS, FOURTH EDITION Larry R. Griffey • Nonpooled t-test statistic for H0 : µ1 H µ2 (independent samples, and normal populations or large samples): x1 − x2 tH q 2 (s1 /n1 ) + (s22 /n2 ) • Sample size for estimating µ: zα/2 · σ 2 , nH E rounded up to the nearest whole number. with df H 1. • Studentized version of the variable x: x−µ tH √ s/ n • Nonpooled t-interval for µ1 − µ2 (independent samples, and normal populations or large samples): q (x 1 − x 2 ) ± tα/2 · (s12 /n1 ) + (s22 /n2 ) • t-interval for µ (σ unknown, normal population or large sample): s x ± tα/2 · √ n with df H 1. • Paired t-test statistic for H0 : µ1 H µ2 (paired sample, and normal differences or large sample): with df H n − 1. CHAPTER 9 Hypothesis Tests for One Population Mean tH • z-test statistic for H0 : µ H µ0 (σ known, normal population or large sample): zH with df H n − 1. • Paired t-interval for µ1 − µ2 (paired sample, and normal differences or large sample): sd d ± tα/2 · √ n with df H n − 1. x − µ0 √ σ/ n • t-test statistic for H0 : µ H µ0 (σ unknown, normal population or large sample): tH d √ sd / n x − µ0 √ s/ n CHAPTER 11 Inferences for Population Proportions • Sample proportion: with df H n − 1. x , n where x denotes the number of successes. p̂ H CHAPTER 10 Inferences for Two Population Means • Pooled sample standard deviation: s (n1 − 1)s12 + (n2 − 1)s22 sp H n1 + n2 − 2 • One-sample z-interval for p: p̂ ± zα/2 · p p̂(1 − p̂)/n (Assumption: both x and n − x are 5 or greater) • Pooled t-test statistic for H0 : µ1 H µ2 (independent samples, normal populations or large samples, and equal population standard deviations): x1 − x2 tH √ sp (1/n1 ) + (1/n2 ) with df H n1 + n2 − 2. • Margin of error for the estimate of p: p E H zα/2 · p̂(1 − p̂)/n • Sample size for estimating p: zα/2 2 zα/2 2 n H 0.25 or n H p̂g (1 − p̂g ) E E rounded up to the nearest whole number (g H “educated guess”) • Pooled t-interval for µ1 − µ2 (independent samples, normal populations or large samples, and equal population standard deviations): p (x 1 − x 2 ) ± tα/2 · sp (1/n1 ) + (1/n2 ) • One-sample z-test statistic for H0 : p H p0 : p̂ − p0 zH √ p0 (1 − p0 )/n (Assumption: both np0 and n(1 − p0 ) are 5 or greater) with df H n1 + n2 − 2. • Pooled sample proportion: p̂p H • Degrees of freedom for nonpooled-t procedures: i2 h s12 /n1 + s22 /n2 1H 2 2 , s12 /n1 s22 /n2 + n1 − 1 n2 − 1 rounded down to the nearest integer. x1 + x2 n1 + n2 • Two-sample z-test statistic for H0 : p1 H p2 : p̂1 − p̂2 zH p √ p̂p (1 − p̂p ) (1/n1 ) + (1/n2 ) (Assumptions: independent samples; x1 , n1 − x1 , x2 , n2 − x2 are all 5 or greater) -2- FORMULA CARD FOR WEISS’S ELEMENTARY STATISTICS, FOURTH EDITION Larry R. Griffey • Two-sample z-interval for p1 − p2 : p (p̂1 − p̂2 ) ± zα/2 · p̂1 (1 − p̂1 )/n1 + p̂2 (1 − p̂2 )/n2 • One-way ANOVA identity: SST H SSTR + SSE • Computing formulas for sums of squares in one-way ANOVA: (Assumptions: independent samples; x1 , n1 − x1 , x2 , n2 − x2 are all 5 or greater) SST H 6x 2 − (6x)2 /n SSTR H 6(Tj2 /nj ) − (6x)2 /n • Margin of error for the estimate of p1 − p2 : p E H zα/2 · p̂1 (1 − p̂1 )/n1 + p̂2 (1 − p̂2 )/n2 SSE H SST − SSTR • Mean squares in one-way ANOVA: • Sample size for estimating p1 − p2 : zα/2 2 n1 H n2 H 0.5 E or 2 z α/2 n1 H n2 H p̂1g (1 − p̂1g ) + p̂2g (1 − p̂2g ) E rounded up to the nearest whole number (g H “educated guess”) MSTR H SSTR , k−1 MSE H SSE n−k • Test statistic for one-way ANOVA (independent samples, normal populations, and equal population standard deviations): F H MSTR MSE with df H (k − 1, n − k). CHAPTER 12 Chi-Square Procedures CHAPTER 14 Inferential Methods in Regression and Correlation • Expected frequencies for a chi-square goodness-of-fit test: E H np • Population regression equation: y H β0 + β1 x r SSE • Standard error of the estimate: se H n−2 • Test statistic for a chi-square goodness-of-fit test: χ 2 H 6(O − E)2 /E with df H k − 1, where k is the number of possible values for the variable under consideration. • Test statistic for H0 : β1 H 0: tH • Expected frequencies for a chi-square independence test: R·C EH n where R H row total and C H column total. with df H n − 2. • Confidence interval for β1 : se b1 ± tα/2 · √ Sxx • Test statistic for a chi-square independence test: with df H n − 2. χ 2 H 6(O − E)2 /E with df H (r − 1)(c − 1), where r and c are the number of possible values for the two variables under consideration. • Confidence interval for the conditional mean of the response variable corresponding to xp : s (xp − 6x/n)2 1 + ŷp ± tα/2 · se n Sxx CHAPTER 13 Analysis of Variance (ANOVA) • Notation in one-way ANOVA: with df H n − 2. k H number of populations n H total number of observations x H mean of all n observations nj H size of sample from Population j x j H mean of sample from Population j sj2 H variance of sample from Population j Tj H sum of sample data from Population j • Prediction interval for an observed value of the response variable corresponding to xp : s (xp − 6x/n)2 1 + ŷp ± tα/2 · se 1 + n Sxx with df H n − 2. • Test statistic for H0 : ρ H 0: • Defining formulas for sums of squares in one-way ANOVA: SST H 6(x − x) b1 √ se / Sxx tH s 2 SSTR H 6nj (x j − x) 2 with df H n − 2. SSE H 6(nj − 1)sj2 -3- r 1 − r2 n−2