Statistical tests inventory

	A	B	C	D	E	F	G	H	I	J	K	N	O	P	Q	S	T	U	V	W
1	Data	ID	statkat.com ID; append to https://statkat.com/stattest.php?&t= to see	statkat.com link	Family	Subtype	Priority	Priority comments	H0	HA	Effect size	Purpose	AKA	Data	Assumptions	Sampling distribution under H0	Formulas (estimators, effect size, power, CIs)	Resampling method equivalent	Nonparametric option	Notes

2
3		X2			Chi square tests		P2
4	categorical	X2GF	3	3		Chi square test for goodness of fit	P2					Check if an experimental distribution matches the theoretical expected			- The sampling method is simple random sampling - The expected frequency count for each cell of the table is at least 5
5	categorical	X2I	4	4		Chi square test of independence	P2		H0: there is no association between the two variables (independent).	HA: there is an association between the two variables.		Check for association between two categorical variables	Chi-squared test for the relationship between two categorical variables	one sample with at least two cateorical variables	- The sampling method is simple random sampling - The expected frequency count for each cell of the table is at least 5
6	categorical	X2H				Chi square test for homogeneity	P2		H0: proportions are the same (p1= p2= p3= …).	HA: at least one proportion is different the stated proportions .		Check whether 2 or more independent random samples are drawn from the same population or from different populations		two samples of one categorical variable from different populations	- For each population, the sampling method is simple random sampling - The expected frequency count for each cell of the table is at least 5					chi square test of homogeneity is an extension of chi square test of independence.
7	numerical	X2VAR				Chi square test for population variance	P3	Including only because already written up in Sec 3.3				Compare a sample variance s2 to a known theoretical variance sigma2
8
9
10
11		Z			Z-Test		P2
12	numerical	Z1	5	5		One sample z-test for the mean	P2	covered in Sec 3.3 as intro example				Compare a sample mean to a theoretical mean, assuming population variance is known		Compare a sample independent observations with theory
13	numerical	Z2	8	8		Two sample z-test for the mean	NO	better cover just t (otherwise too much)				Compare difference of sample means, assuming population variance is known		Compares two independent samples
14
15
16
17		PROP			Proportion tests		P2
18	categorical (binary)	PROP1	1	1		z-test for a single proportion	P2
19	categorical (binary)	BINOM	44	44		Binomial test for a single proportion	P3
20	categorical (binary)	PROP2	2	2		z-test for the difference between two proportions	P2
21
22
23		T			T-Tests		P1						Student's t-test
24	numerical	T1	6	6		One sample t-test	P2							Compare a sample independent observations with theory	- n greater than 20				1-sample sign; 1-sample Wilcoxon
25	numerical	T2W	9	9		Welch's t-test	P1	Primary					Welch's unequal variances t-test	Compares two independent samples from populations with differet variance
26	numerical	T2P	10	10		Two-sample t-test	P2	Do not focus on this, since T2W is always better				pooled variance estimate		Compares two independent samples	- each group should have n greater than 15				Mann-Whitney test
27	numerical	T1p	7	7		Paired t-test	P2							Examines a set of differences
28
29
30		ANOVA			ANOVA		P2						Analysis of Variance, F-test
31	numerical	ANOVA	11	11		One-way ANOVA	P2							Compares the menas of three or more groups	- each group should be greater than 15				Kruskal-Wallis; Mood's median test
32		ANOVA2W	12	12		Two-way ANOVA	P3						Factorial ANOVE with two factors	Compares groups classified by two different factors					Friedman test
33					Tukey			Tukey HSD? super important if you're covering post-hoc!						Perform pairwise comparison between all groups
34		ANOVArm				One-way repeated measures ANOVA	P99						One-way ANOVA paired test
35
36		NP			Nonparametric		P3
37	numerical or ordinal	SIGN				Sign test for a population median	P3							weaker version of Wilcoxon signed rank test that looks only at sign and not rank
38		SIGNP	34	34		Sign test for matched-pairs	P99
39
40	numerical or ordinal	WSIGNEDR	42	42		One sample Wilcoxon signed-rank test	P2
41		WSIGNEDRP	35	35		Wilcoxon signed-rank test for matched-pairs	P99
42	numerical or ordinal	MWU	14	14		Mann-Whitney U-test	P3						Wilcoxon rank sum test, Unpaired two-samples Wilcoxon test
43	numerical or ordinal	KW	17	17		Kruskal-Wallis analysis of variance by ranks	P3
44		FRI	33	33	Friendman test		?
45		FE			Fisher's exact test	in R	?					on biostathandbook	see also Barnard's test	2x2 contingency table
46
47
48		REG			Regression		P1						Analysis of Variance, F-test
49	numerical	REGT				Regression coefficient test?	P3	this is pretty similar in process to a one-sample t-test!				test for correlation coeff		Measure the linear association between two variables
50	numerical	REG	13	13		Regression (OLS)	P1	Chapter 4				test for y-intercept or slope (any regression coefficient)
51	numerical	REGF				Regression F test (equivalent to ANOVA?)	P2	Chapter 4
52	categorical	LOGREG	20	20		Logistic regression	P3					logit(LinearRegression)
53	numerical	MREG	25	25		Multiple linear regression	P3					LogisticRegression with multiple input variables
54		MLOGREG	21	21		Multinomial logistic regression	P99					special case of the above, via pre-computations (design matrix)
55		OLOGREG	22	22		Ordinal logistic regression	P99	this might be more important for business people, because I see WAY TOO MANY people treating their ordinal variables as continuous...its one of the most common handful of consults I get												https://www.statsmodels.org/stable/examples/notebooks/generated/ordinal_regression.html
56
57
58																			Spearman Rank Correlation
59
60		CORR			Correlation		P3
61	numerical	PEAR	19	19		Pearson correlation	P2
62	numerical or ordinal	SPEAR	18	18		Spearman rho	P3	Chapter 4
63
64
65					Distribution checks
66	numerical	KS				Kolmogorov-Smirnov Test	P3	(Not very good: accepts all small samples, rejects large samples)
67	numerical	SW				Shapiro-Wilk test for normality	P3
68		AP				D’Agostino and Pearson’s test based on skew and kurtosis	P99
69		AD				Anderson-Darling test for data coming from a particular distribution	P99
70	numerical	LEV				Levene's test for homogeneity of variance	P99
71
72
73
74
75		MANOVA1W	15	15		Multivariate analysis of variance	P99
76		MANOVA2W	16	16			P99
77
78		ANCOVA	23	23		ANCOVA	P99					Linear regression with a mix of caregorical and continuous variables
79		MANCOVA	24	24		MANCOVA	P99
80		ANOVAP	28	28		Repeated measures ANOVA	P99						RMA
81			29	29		Repeated measures ANCOVA	P99
82		?P	39	39		McNemar's test	P99
83		?P	40	40		Cochran's Q test	P99
84						Mood's median test	P99
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100