N318b Winter 2002 Nursing Statistics Specific statistical tests Chi-square (  2 ) Lecture 7.

Slides:



Advertisements
Similar presentations
Hypothesis Testing Steps in Hypothesis Testing:
Advertisements

Chi square.  Non-parametric test that’s useful when your sample violates the assumptions about normality required by other tests ◦ All other tests we’ve.
Chi Square Tests Chapter 17.
INTRODUCTION TO NON-PARAMETRIC ANALYSES CHI SQUARE ANALYSIS.
Analysis of frequency counts with Chi square
1-1 Copyright © 2015, 2010, 2007 Pearson Education, Inc. Chapter 25, Slide 1 Chapter 25 Comparing Counts.
Making Inferences for Associations Between Categorical Variables: Chi Square Chapter 12 Reading Assignment pp ; 485.
PY 427 Statistics 1Fall 2006 Kin Ching Kong, Ph.D Lecture 12 Chicago School of Professional Psychology.
CJ 526 Statistical Analysis in Criminal Justice
Chapter Goals After completing this chapter, you should be able to:
Chi Square Test Dealing with categorical dependant variable.
Chi-square Test of Independence
Ch 15 - Chi-square Nonparametric Methods: Chi-Square Applications
PSY 307 – Statistics for the Behavioral Sciences Chapter 19 – Chi-Square Test for Qualitative Data Chapter 21 – Deciding Which Test to Use.
Today Concepts underlying inferential statistics
1 Nominal Data Greg C Elvers. 2 Parametric Statistics The inferential statistics that we have discussed, such as t and ANOVA, are parametric statistics.
Chapter 12 Inferential Statistics Gay, Mills, and Airasian
The Chi-square Statistic. Goodness of fit 0 This test is used to decide whether there is any difference between the observed (experimental) value and.
1 of 27 PSYC 4310/6310 Advanced Experimental Methods and Statistics © 2013, Michael Kalsher Michael J. Kalsher Department of Cognitive Science Adv. Experimental.
AM Recitation 2/10/11.
CHP400: Community Health Program - lI Research Methodology. Data analysis Hypothesis testing Statistical Inference test t-test and 22 Test of Significance.
N318b Winter 2002 Nursing Statistics Specific statistical tests: Correlation Lecture 10.
1 Psych 5500/6500 Chi-Square (Part Two) Test for Association Fall, 2008.
Single-Sample T-Test Quantitative Methods in HPELS 440:210.
Education 793 Class Notes T-tests 29 October 2003.
Copyright © 2010, 2007, 2004 Pearson Education, Inc. Chapter 26 Comparing Counts.
Copyright © 2010 Pearson Education, Inc. Warm Up- Good Morning! If all the values of a data set are the same, all of the following must equal zero except.
NONPARAMETRIC STATISTICS
CJ 526 Statistical Analysis in Criminal Justice
Comparing Two Population Means
N318b Winter 2002 Nursing Statistics Hypothesis and Inference tests, Type I and II errors, p-values, Confidence Intervals Lecture 5.
Chi-square (χ 2 ) Fenster Chi-Square Chi-Square χ 2 Chi-Square χ 2 Tests of Statistical Significance for Nominal Level Data (Note: can also be used for.
Chapter 9: Non-parametric Tests n Parametric vs Non-parametric n Chi-Square –1 way –2 way.
N318b Winter 2002 Nursing Statistics Lecture 2: Measures of Central Tendency and Variability.
FOUNDATIONS OF NURSING RESEARCH Sixth Edition CHAPTER Copyright ©2012 by Pearson Education, Inc. All rights reserved. Foundations of Nursing Research,
Chapter 16 The Chi-Square Statistic
N318b Winter 2002 Nursing Statistics Specific statistical tests: Regression Lecture 11.
Education 793 Class Notes Presentation 10 Chi-Square Tests and One-Way ANOVA.
Slide 26-1 Copyright © 2004 Pearson Education, Inc.
Nonparametric Tests: Chi Square   Lesson 16. Parametric vs. Nonparametric Tests n Parametric hypothesis test about population parameter (  or  2.
Educational Research Chapter 13 Inferential Statistics Gay, Mills, and Airasian 10 th Edition.
+ Chi Square Test Homogeneity or Independence( Association)
CHI SQUARE TESTS.
HYPOTHESIS TESTING BETWEEN TWO OR MORE CATEGORICAL VARIABLES The Chi-Square Distribution and Test for Independence.
Copyright © 2010 Pearson Education, Inc. Slide
Experimental Research Methods in Language Learning Chapter 10 Inferential Statistics.
Inferential Statistics. The Logic of Inferential Statistics Makes inferences about a population from a sample Makes inferences about a population from.
Chapter 13 Inference for Counts: Chi-Square Tests © 2011 Pearson Education, Inc. 1 Business Statistics: A First Course.
Chapter Outline Goodness of Fit test Test of Independence.
Copyright © 2010 Pearson Education, Inc. Warm Up- Good Morning! If all the values of a data set are the same, all of the following must equal zero except.
Non-parametric Tests e.g., Chi-Square. When to use various statistics n Parametric n Interval or ratio data n Name parametric tests we covered Tuesday.
Chapter 15 The Chi-Square Statistic: Tests for Goodness of Fit and Independence PowerPoint Lecture Slides Essentials of Statistics for the Behavioral.
Bullied as a child? Are you tall or short? 6’ 4” 5’ 10” 4’ 2’ 4”
Chapter 14 – 1 Chi-Square Chi-Square as a Statistical Test Statistical Independence Hypothesis Testing with Chi-Square The Assumptions Stating the Research.
1 1 Slide © 2008 Thomson South-Western. All Rights Reserved Chapter 12 Tests of Goodness of Fit and Independence n Goodness of Fit Test: A Multinomial.
Comparing Counts Chapter 26. Goodness-of-Fit A test of whether the distribution of counts in one categorical variable matches the distribution predicted.
T-tests Chi-square Seminar 7. The previous week… We examined the z-test and one-sample t-test. Psychologists seldom use them, but they are useful to understand.
N318b Winter 2002 Nursing Statistics Specific statistical tests: Tests for means when there are more than 2 groups ANOVA Lecture 9.
N318b Winter 2002 Nursing Statistics Specific statistical tests: The T-test for means Lecture 8.
Educational Research Inferential Statistics Chapter th Chapter 12- 8th Gay and Airasian.
Chapter 11: Categorical Data n Chi-square goodness of fit test allows us to examine a single distribution of a categorical variable in a population. n.
Statistical principles: the normal distribution and methods of testing Or, “Explaining the arrangement of things”
Cross Tabulation with Chi Square
Chapter 9: Non-parametric Tests
Lecture Nine - Twelve Tests of Significance.
INTRODUCTORY STATISTICS FOR CRIMINAL JUSTICE
Qualitative data – tests of association
The Chi-Square Distribution and Test for Independence
Chapter 18: The Chi-Square Statistic
Chapter 26 Comparing Counts.
Presentation transcript:

N318b Winter 2002 Nursing Statistics Specific statistical tests Chi-square (  2 ) Lecture 7

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 2 Today’s Class  5 basic statistical tests covered in course  Parametric and non-parametric tests  Degrees of freedom  >  Example of chi-square test  Applying knowledge to assigned readings Turk et al. (1995) Followed by small groups 12-2 PM Focus on interpreting chi-square results

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 3 “In Group” Session Focuses on an assigned reading. Q1 example of the chi square test Q2 example of the chi square test Q3 criteria for non-parametric test Key points from the Turk et al paper will be covered in the 2 nd part of the lecture Missing Table 1

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 4 New Lecture Material Specific statistical tests: Parametric and non- parametric tests

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 5 Specific Statistical Tests Course will cover five major “tests”: 1. Chi-square (  2 ) 2. T-tests 3. Analysis of variance (ANOVA) 4. Correlation 5. Regression

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 6 Statistical Tests – cont’d All these tests do basically the same 3 things: 3. “test statistic” follows known distributions such that the probability of its value occurring can be determined (i.e. its “p-value”) 2. Generate a “test statistic” whose value increases as difference between groups increases (i.e. larger values more significant) 1. Compare 2 or more study groups to each other (or one group to a reference group) Example: Z-scores

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 7 Statistical Tests – cont’d How do you known when to use which test? Helps to ask some basic questions: 1. What kind of data are used? 2. What kind of relationship is of interest? 3. How many groups (samples) involved? - one, two, or more than two - prediction, association or difference? - ratio/interval or categorical (ordinal/nominal) - dependent (e.g. follow-up) or independent

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 8 Key point is determining type of data For categorical (i.e. either nominal or ordinal data) the normal distribution is generally not applicable and population descriptors (parameters) cannot be estimated so non-parametric tests used Main non-parametric test is the chi-square test that compares expected (E) numbers with actual or observed (O) numbers Non-Parametric Tests

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 9 For continuous (i.e. either interval or ratio data) the normal distribution applies and population descriptors (parameters, like means) can be estimated thus parametric tests are used instead Main tests for this course include the t-test, paired t-test and analysis of variance (ANOVA), all of which test means Parametric Tests

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 10 Parametric vs. non- parametric tests Data usedExamplesComments Non- parametric (numbers, %’s) Nominal, ordinal (categorical) Chi- square Easy to use but limited to simple situations Parametric (means, variances) Interval, ratio (continuous) T-tests, ANOVA, regression More flexible and powerful (also more convincing)

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 11 Degrees of Freedom SD =  (x-  ) 2 n -1 Recall the formula for SD was “adjusted” for imprecision of small samples The (n-1) term is referred to as “degrees of freedom” since it indicates how many ways that the data can vary in a sample

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 12 Degrees of Freedom – cont’d Value of “test statistic” derived from many statistical tests is dependent on this idea of “degrees of freedom” thus some sense of what it means is useful (e.g. see textbook page 84-85) df = number of ways that data can vary (or be categorized) Example – for chi square test: df = (number of categories –1)

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 13 Example – for chi square test: df = (number of categories –1) Degrees of Freedom – cont’d Why? If total number of subjects is known, and they are categorized into 4 groups, then if three tallies are known the fourth is “fixed” – i.e. it can be derived so it is not “free” to vary df =(4 –1) = 3

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 14 Chi square (  2 ) test How do you known when to use  2 test? 1. What kind of data are used? 2. What kind of relationship is of interest? 3. How many groups (samples) involved? - categorical ( typically nominal) - frequencies (i.e. counts or percentages) - data can be put in a “contingency table” - association or difference - usually two or more (“smallish” number) Referring back to the 3 “basic questions”:

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 15 Chi square test - example One of the most common statistical tests ! Example: We suspect that students at UWO love statistics a lot so we ask 100 nursing students if they really like Nur 318b? 63 say YES, 37 say NO Is this more than we might have expected – i.e. are UWO nurses crazy about statistics?

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 16 If we did not think students would be more or less likely to enjoy the course, we would EXPECT 50 to say no and 50 to say YES Chi square test - example  2 compares observed vs expected numbers H 0 : no difference in OBS versus EXP counts H a : OBS count is NOT equal to EXP Study hypotheses

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 17 Chi square test - example YESNO at UWO (observed) 6733 In general (expected) 50 22 =  (O-E) 2 E (67-50) 2 + (33-50) 2 50 == 11.56

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 18 Chi square test - example As with Z-scores, we now look this number (11.56) up in a table of critical values, in this case for the chi square distribution (table value is the probability that observed and expected numbers are the same) 22 (1 df) = 11.56, p < Thus we can conclude that UWO nursing students must love stats !!!

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page minute break !

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 20 Chi square test - assumptions 1. Data are counts, frequencies, percentages 2. Smallest table cell counts ideally >5 3. Data in rows and columns are independent (i.e. subjects can be in one table cell only) 4. Categories or levels set BEFORE testing

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 21 Why is the chi square a nonparametric statistical test? Chi square test - assumptions 1) it does not assume data are normally distributed (in fact NO assumptions are needed about underlying distribution) 2) categorical/nominal data are used 3) not estimating a population characteristic (i.e. a parameter, like the mean)

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 22 Part 2: Application to the Assigned Readings

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 23 Turk et al. (1995) Quick summary of the paper: – a cross-sectional study examining the cognitive-behavioral mediation model of depression in chronic pain patients – 100 chronic pain subjects divided into two groups: 73 randomly chosen younger (<70); and 27 older (  70 yrs) patients – found a strong link between pain and depression for older subjects but not for younger ones (i.e. an age effect)

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 24 Some design issues? Do you have any concerns with design of the study – e.g. using a cross-sectional design to examine chronic pain and depression? Can pain be more of “social” problem with older people thus “confounding” assessment of depression? Which came first (“chicken-and-egg”)? Was assessment of depression “blinded”?

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 25 Chi square test – example 2: the contingency table GenderyoungoldTotal Male45.21 (33) (10)43 Female54.79 (40) (17)57 Total100% (73) 100% (27)100 Observed counts from Table 1

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 26 Chi square test – example 2: the contingency table How did we get counts from %’s? Just multiply % by total number in group e.g % male in younger group is equal to x 73 = 33 males How do we get expected counts? Expected counts assume no association between groups thus they are calculated according to size of cells in groups

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 27  2 Contingency Table Expected counts R i x C j N E ij = = For cell 1,1: R 1 x C E 11 = 43 x =31.4 For cell 1,2 = 11.6 For cell 2,1 = 41.6 For cell 2,2 = 15.4

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 28 GenderyoungoldTotal Male33 (31.4) 10 (11.6)43 Female40 (41.6) 17 (15.4)57 Total  2 Contingency Table Expected counts C1C1 C2C2 R1R1 R2R2

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 29  2 Contingency Table Test statistic 22 =  (O-E) 2 E ( ) 2 + ( ) = ( ) 2 + ( ) 22 (1 df) = 0.54, p > 0.20 Can’t reject null hypothesis, thus no association !

School of Nursing Institute for Work & Health Nur 318b 2002 Lecture 6: page 30 Next Week - Lecture 8: T-test For next week’s class please review: 1.Page 16 in syllabus 2.Textbook Chapter 4, pages Syllabus paper: Turk et al. (1995)