Quiz 5 quiz STATISTICAL TESTS
what do you use for type 2 error? a. alpha values b. beta values c. gamma values
b
what kind of testing are type 1 & 2 errors related to? a. statistical testing b. analysis testing c. hypothesis testing d. reliant testing
c
Study data were analyzed using the SPSS statistical package. Tests applied tot he data included the levee test for equality of variances to determine whether equal variance between groups could be assumed. equal variances between groups support use of conventional statistical analysis, whereas without this assumption, transformation of the data may be required prior to making comparisons. Independent samples t tests and pared t tests allowed for comparisons between the pretreatment and posttreatment test results between and within groups, respectively. the Mann-Whitney rank sure test was used to compare test results between groups for ordinal data such as berg balance scores. the Spearmen rho test for nonparametric data was used for correlation analysis between groups and between measures. priori, significance was set at p less than .05 1. what assesses if equal variance could be assumed? 2. what does the independent samples t tests compare? 3. what does the paired t tests compare? 4. what kind of tests is Mann-whitney rank and the Spearman rho test assess? a. parametric b. nonparametric
1. levene test 2. between groups 3. within groups 4. b (ordinal data)
We are looking to compare the outcomes of older adults with balance disorders treated in 3 clinics each of which plays different music in the background clinic 1- no music clinic 2- classical music clinic 3- rap music on timed one legged standing 1. what is the IV 2. what is the DV 3. what kind of data 4. how many factors 5. how many levels
1. music in the background 2. timed one leg stand 3. IV data 4. 1 5. 3
when looking at sum of squares, we want to explain how much variability is due to the effect of our experiment (1) and how much variability is due to unexplained sources (error) (2)... which corresponds -between group -within group
1.between group 2.within group (error)
in my musical experiment... I get a SStotalvariability of 4200 and a SSbetweengroups of 3200... what would be the SSerror (within groups)?
1000
what is the most often seen nonparametric test in the PTJ?
Chi squared
I have the mean squares of between group variance (1600) and the mean square of within group variance (111.11).... what am I doing to get the f value?
MSbetween/MSerror
Title: balance and mobility following stroke: effects of PT interventions with and without biofeedback/ force plate training methods: -random assignment - Group 1: PT interventions designed to improve balance and mobility 2-3 times per week for 4 wks - Group 2: PT interventions designed to improve balance and mobility 2-3 times per wk plus neurocom balance master for 15 min of each treatment session for 4 wks - outcome measures: berg balance scale what's the P I C O
P- outpatients past stroke I- neurocom balance master C- no neurocom O- balance and mobility
Alli does an ANOVA test and decides that she is going to measure the total variance by lumping all subjects together. She then measures the variance of each group separately. Strangely enough, the total group variation and the within subject variation is extremely close. what does that mean for between group variance? a. small, as a lot of overlap occurs b. small, as distribution is close c. large, as a lot of overlap occurs d. large, as distribution is close
a
Dakota has an F statistic of 14.400 and gets a p value of .002.. he has a standard alpha value that is related to a 95% CI... what is the determination from this? a. reject the null b. fail to reject the null c. prove the null d. cannot be determined
a
Dakota is doing a study that he has a ton of time and money invested into. His results are ify but he is so excited about it he concluded that there was a real difference in samples when it was entirely due to chance :( travis does not know really what hes looking at with research, reads this released doc and begins to used this treatment in his clinic, even though it is not ACTUALLY effective. this is an example of: a. type 1 error b. type 2 error c. standard error
a
Dakota takes two samples of people with chronic shoulder pain and splits them into groups. He assigns one to have a joint mobility program while the other receives a joint mobility program and ultrasound. Based on the data that he recieves... he finds that the JM+ultrasound group has made signicantly greater recovery than the other group... in this case he would a. reject the null b. fail to reject the null c. eject the null
a
Katie determines that there is no difference between strength in genders.... this is a _____ a. null hypothesis b. alternative hypothesis c. theory hypothesis d. if you are here go up this aint the answer
a
Statistical tests evaluate the plausibility of a particular hypothesis about populations taking into account the results observed in the _____ a. samples b. point estimate c. target population d. sampling frame
a
Using a chi square... I am looking for whether or not frequencies differ from distribution this is an example of a. goodness of fit b. test independence
a
When determining significance level.... the criterion used to evaluate the differences between what we expect on the basis of hypothesis is the ____ a. p-value b. alpha value c. mean of means
a
true of false p is the probability of the null hypothesis
false
two means are equal a. null b. one sided alternative hypothesis (directional) c. two sided alternative hypothesis
a
used to compare two means What kind of test: a. t test b. anova c. z test d. chi squared
a
we know: statistical significance DOES NOT imply meaningful or clinical significance. what alteration makes many relationships statistically significant? a. larger samples b. smaller samples c. cluster samples d. subgroup splitting
a
what do you use for type 1 error? a. alpha values b. beta values c. gamma values
a
what is the result of dividing the sum of squares by the degrees of freedom in a between group variance? a. mean square (MS) b. f value c. p value d. t value
a
what is this characterizing: characterized by the fact that we act as if hypothesis is true unless it is clearly contradicted by the sample data. a. null b. alternative
a
what must be assumed with probability values? a. null is true b. null is false c. null is irrelevant
a
when is turkeys HSD perferred a. large groups b. small groups c. all the time d. one group
a
when looking at a t-test, what does increased variability in denominator mean? a. more error b. less error c. more difference d. less difference
a
which post hoc analysis: a. turkeys honestly significant difference b. newman-keuls method c. sheffe test d. bonferroni correction I wanted to look at the types of music in a clinic and the effect it would have on participation. I looked at rap, country, folk, pop, r&b, jazz, and reggae. I was conservative in my pairwise comparison as I wanted to ensure I was not making type 1 error. I wanted to see all differences. I kept the same critical value for each comparison
a
which test wants to make sure that I do not mistakenly say there is an effect when there isn't one in my anova and post hoc a. conservative test b. liberal test
a
which would be most appropriate in these cases a. nonparametric b. parametric nominal level data
a
which would be most appropriate in these cases a. nonparametric b. parametric normality can't be assumed
a
which would be most appropriate in these cases a. nonparametric b. parametric ordinal level data
a
with chi squared tests... there is time where assumptions of frequency or cell sizes are not met. in that case you need a continuity correction. which one will I need if the expected frequency in any of the cells of my 2x2 is less than 5? a. yates correction b fishers exact test
a
Typically P values range ____ a. 0-1 b. 0-10 c. 0-100 d. -50-50
a (% probability that difference of statistic being tested (ie mean) occurred by chance)
for CHI SQUARE TESTS the null hypotheses is p1=p2=pk as you are looking at the proportions of a sample.. it is used when you have what kind of data? a. nominal (frequency data) b. ordinal data c. IR data
a (MAKES SENSE... nominal data does not follow skew and becomes a nonparametric test)
We are looking to compare the outcomes of older adults with balance disorders treated in 3 clinics each of which plays different music in the background clinic 1- no music clinic 2- classical music clinic 3- rap music What is I called if I am looking at how much difference between no music and classical
between group variance
Dakota has an F statistic of 14.400 and gets a p value of .002.. how did he get this? a. dividing F stat by MS b. square rooting F stat c. seeing where F stat is in relation to its distribution d. removing error in F stat
c
I have a sample of men and women who do and dont exercise question: is gender associated with exercise? drawing a two by two I fill in my observed values. ME-20 FE-60 MDE-40 FDE- 80 total males-60 total females- 140 exercisers- 80 non exercisers-120 expected value of exercisers is .4 or 40%... with this information... what would be my expected value for number of females who exercise? a. 60(.4)=24 b. 200 (.4)=80 c. 140(.4)=56 d. 120(40)=4800
c
In a Chi Squared test, you are comparing an actual number (frequency) in each group with an expected number that is calculated... which of these is the expected number NOT based on? a. theory b. experience c. randomization d. comparison groups
c
We are looking to compare the outcomes of older adults with balance disorders treated in 3 clinics each of which plays different music in the background clinic 1- no music clinic 2- classical music on timed one legged standing what kind of test? a. one way anova b. two way anova c. t test
c
What is produced by dividing each sum of squares by its respective degrees of freedom to produce an estimate variance? a. SS b. df c. MS d. F e. p
c
X1=X2=X3=Xk... this is what? a. null hypothesis of t tests b. alternative hypothesis of t tests c. null hypothesis of ANOVA d. alternative hypothesis of ANOVA
c
a standard alpha of .05 is equivalent to what CI? a. 50% b. 75% c. 95% d. 99%
c
alpha level defines the maximal acceptable risk of making a _____ if you reject the null a. standard error b. sampling error c. type 1 error d. type 2 error
c
if within group variation and total group variations are equal then... a. you have little variation between groups b. you have large variation between groups c. you have no variation between groups
c
null hypothesis in z tests are comparing? a. sample data b. sample means c. population means d. population estimates
c
one step of hypothesis testing is to state the null and assume it to be true until testing is completed. The difference between the statistic computed in the sample and the parameter specified by null hypothesis is computed..... this difference is known as a. t score b. z score c. p value d. ANOVA
c
travis, like the jackass he is, gets really confused when he sees data. In this one study he did, he even concluded that the null hypothesis was not true when in fact it was... this is an example of? a. type 1 error b. type 2 error c. standard error
a
We are looking to compare the outcomes of older adults with balance disorders treated in 3 clinics each of which plays different music in the background clinic 1- no music clinic 2- classical music clinic 3- rap music on timed one legged standing what kind of test? a. one way anova b. two way anova c. t test
a
You see an ANOVA in which the groups do not overlap. Distance between each group is greater than the distance within each group. a. statistically significant ANOVA, large F ratio b. statistically significant ANOVA, small F ratio c. not statistically significant anova , large F ratio d. not statistically significant anova , small F ratio
a
a p-value tells you the percent probability that the ______ occurred by chance a. difference between the stat being tested (ie means) b. similarities between the stat being tested (ie means) c. variation between the stat being tested (ie means)
a
in post hoc tests (multiple comparison tests) each test rank orders means and then tests pairwise differences between means against a critical value. Each test offers a differently level of protection agains Type I and Type II error... which kind of tests is more type I protective, in which means need to be far apart to establish significance? a. conservative test b. liberal test
a
parametric tests are typically used with a. IR data b. ordinal data c. nominal data d. b&c
a
the LEVEL OF SIGNIFICANCE is the value the researcher has selected for rejecting the null hypothesis... or the ______ value a. alpha b. beta c. gamma
a
Katie decides prior to testing (a priori) her hypothesis that her alpha is .05... at conclusion of the analysis, it is determined that her p value is .001 in this case---- what is the max RISK of committing type 1 error? a. .05 b. .001 c. n-1
a (just cuz researcher got a p value lower does not mean the risk is any smaller)
which would be most appropriate in these cases a. nonparametric b. parametric data with small sample sizes
a (not normally distributed)
which p value leads to a less plausible null and more faith in a difference a. .001 b. .05 c. .5 d. 10
a (smaller p value)
Frazier decides prior to testing (a priori) her hypothesis that her alpha is .05... at conclusion of the analysis, it is determined that her p value is .001. what would she do with this info? a. reject null b. fail to reject null c. prove null
a (there may really be a difference)
what would cause IR data to be non parametric? a. if variable is not normally distributed in population b. if variable is not normally distributed in sample c. if variances in sample are equal d. b&c
a (think parametric tests are assumptions: variable should be normally distributed in population and variances in sample are equal)
Alli decides to reject the null hypothesis based on the fact that her data shows it is unlikely that chance is producing the observed differences.... this means that the effect is? a. void b. significant c. negligable d. pointless
b
Dakota sees that his p is above the alpha (significance level), what does that mean a. reject the null b. fail to reject the null c. the null is proven
b
Emily obtains a p value of .005 with an association test statistic... what does it mean? a. the probability that the null hypothesis is true is .005 b. the probability of obtaining data as different of more different from the null hypothesis as those obtained in the experiment is .005 c. the probability of me failing this test is so likely that I am already applying to several fast food jobs d. the probability of us killing it on this test is .005
b
I have a sample of men and women who do and dont exercise question: is gender associated with exercise? drawing a two by two I fill in my observed values. ME-20 FE-60 MDE-40 FDE- 80 total males-60 total females- 140 exercisers- 80 non exercisers-120 I determine that chi squared value through summing my (observed minus expected)^2 /expected and determining my degrees of freedom... which gives me an x^2=1.587. I look at my p value associated and see p=.208 I am unable to reject the null.... therefore I conclude that? a. there is an association between gender and exercise.. women win b. there is no association between gender and exercise c. there is an association between gender and exercise.. men win d. nothing can be concluded
b
I have a sample of men and women who do and dont exercise question: is gender associated with exercise? if gender were not associated to exercise... what would we expect the proportions of men and women who exercise to be in our chi square? a. similar b. equal c. different d. cannot be determined
b
If total group variance is much larger than average variances within the separate groups then you have a significant mean difference between ____ a. one group b. at least two groups c. at least three groups d. all groups present
b
Morgan decides prior to testing (a priori) her hypothesis that her alpha is .05... at conclusion of the analysis, it is determined that her p value is .15. what would she do with this info? a. reject null b. fail to reject null c. prove null
b
Morgan wants to make sure her name is not trashed in the real world and works really hard to not falsely publish difference and create bad practice in the clinical setting. In this process, she is a little too careful and concludes there was no real difference between samples when there really was one (fail to reject null) subsequently, alli reads this research and decides that this treatment would not benefit her patients (which sucks because it actually would be effective) This is an example of: a. type 1 error b. type 2 error c. standard error
b
P-value is the probability that differences could have happened by ___ a. a justified cause b. chance c. luck d. random
b
The larger the F statistic... a. the lesser the difference between group means relative to variability within groups b. the greater the difference between group means relative to variability within groups
b
The researchers investigated whether the addition of visual biofeedback/forceplate training could enhance the effects of other PT interventions on balance and mobility. -the sample of convenience of 13 outpatients w hemiplegia who ranged from 30-77 years old the data is? a. super homogenous b. super heterogenous c. perfectly representative of the population
b
We are looking to compare the outcomes of older adults with balance disorders treated in 3 clinics each of which plays different music in the background clinic 1- no music clinic 2- classical music clinic 3- rap music and talking in the Clinic (yes/ no) on timed one legged standing what kind of test? a. one way anova b. two way anova c. t test
b
We know that a null hypothesis states that there is no difference or relationship between variables of interest.... in this case an observed differences are? a. sampling error b. due to chance c. due to regression d. voided
b
What quantifies the amount of variation of scores around the mean for total variation, within group variation, and between group variation in ANOVA a. mean differences b. sum of squares c. difference of squares d. standard deviation
b
What tests is labeled as a binomial nonparametric test a. a test with a z score used for populations through looking population means b. a test with a z score used for population proportions c. a test with a t score looking at sample means d. a test looking at an f score of multiple groups
b
When reading an article, alli looks at statistical difference and sees that it is significant... what does she do with this data a. apply it to clinic as this is good enough information b. be cautious... statistical significance does not mean clinical significance c. get funding and test for herself d. not really care about the data
b
Z tests are typically driven by population data. The numerator: difference between sample means and denominator: SE of sample differences in the formula. What z value relates to an a=.05 and would would you do with the null in this case? a. 1.96, fail to reject b. 1.96, reject c. 2.14, fail to reject d. 2.14, reject
b
______ are used when estimating at least on population parameter from sample statistics assuming that the variable is normally distributed in the population and variances in sample are equal a. parabolic testing b. parametric testing c. nonparametric testing d. anova testing
b
if comparing 4 groups (A,B,C,D) in t tests... there would be 6 analysis that would be done -AvB -AvC -AvD -BvC -BvD -CvD what is the true issue with doing 6 individual t tests a. time consuming b. error would be amplified c. it gets confusing d. you would have no idea where the difference occurs
b
in post hoc tests (multiple comparison tests) each test rank orders means and then tests pairwise differences between means against a critical value. Each test offers a differently level of protection agains Type I and Type II error... which kind of tests is more type II protective, in which means need to be closer to establish significance? a. conservative test b. liberal test
b
most multiple comparison test (post hoc) involve what? a. random ordering b. rank ordering c. stratified order d. place order
b
null hypothesis in Student's t tests are comparing? a. sample data b. sample means c. population means d. population estimates
b
one mean is greater than or less than the other... looking at extremes a. null b. one sided alternative hypothesis (directional) c. two sided alternative hypothesis
b
the probability value is the probability of obtaining data that is ____ than the current data a. that is the same b. as extreme or more extreme c. less extreme d. similar to
b
travis, like the jackass he is, gets really confused when he sees data. In this one study he did, he even concluded that the null hypothesis was true when in fact it was not... this is an example of? a. type 1 error b. type 2 error c. standard error
b
used to compare 3 or more means What kind of test: a. t test b. anova c. z test d. chi squared
b
using chi square.... I am looking to see if two categories of variables independent of eachother+ what is degree of association this is an example of a. goodness of fit b. test independence
b
what does the P value stand for a. percentage b. probability c. progession d. pain
b
what explains total variability that happens in a subgroup? a. SSbetween groups b. SSwithin groups c. SS total variability
b
what is characterizing: regarded as true only if the null is clearly contradicted in the sample a. null b. alternative
b
what is it called ANOVA (analysis of VARIANCE) a. because it is multiple variables b. because it is analyzing variability to see if measures of central tendency differ c. because it is analyzing variability to see if measures of SD differ d. because you obviously aren't analyzing similarities in this case... you wanna see what's different duhhhh
b
what is the most likely group for the t test to be significant a. smaller difference between two means b. larger difference between two means c. no difference between two means
b
what is the problem with statistical tests in relation to the fact that they tell the difference between populations? a. we do not know new population b. we do not know size of difference c. we do not know variance or SE d. there are not CIs
b
what is the purpose of null hypothesis testing a. to find no difference b. determine if null is false c. determine if alternative is false
b
what is the standard alpha that correlates to a 99% CI? a. .005 b. .01 c. .05 d. .1
b
when looking at a t-test, what does increased sample size in denominator mean? a. more error b. less error c. more difference d. less difference
b
when looking at theN SSbetween-group variations.... it is obvious that we are looking at the variation of each group from the grand mean weighted by what? a. total n b. n in each group c. n-1 d. n-2
b
when rejecting the null hypothesis, you are stating..... a. it is likely that chance is producing observed effects b. it is unlikely that chance is producing observed effects c. there is standard error d. i can prove causality
b
when would ordinal data take shape to being a parametric test? a. always b. summating ordinal data.. as it begins to take normal distribution c. squaring ordinal data.. as it begins to take normal distribution d. can't be done its always nonparametric
b
which post hoc analysis: a. turkeys honestly significant difference b. newman-keuls method c. sheffe test d. bonferroni correction I wanted to look at the types of music in a clinic and the effect it would have on participation. I looked at rap, country, folk, pop, r&b, jazz, and reggae. I only want to know when partially evaluate the hypothesis and look at the rap, jazz, and pop as their means seem to be near eachother
b
which test wants to make sure I do not miss an effect and claim there to be no differences in my post hoc/ANOVA a. conservative test b. liberal test
b
which would be most appropriate in these cases a. nonparametric b. parametric the data can study effects of IV on DV and interactions
b
which would be most appropriate in these cases a. nonparametric b. parametric the data is more powerful and flexible
b
with a student's t tests (sample means) and chi squared (proportions of sample means), what influences the distribution? a. variance b. degrees of freedom c. alterations of data d. standard error
b
with chi squared tests... there is time where assumptions of frequency or cell sizes are not met. in that case you need a continuity correction. which one will I need if the sample sizes and expected frequencies are small... aka minimum expected values is less than 5. a. yates correction b fishers exact test
b
what are two statistical uses of chi squared? a. goodness of fit and interdependance b. goodness of fit and test of independance c. relativity and proportion d. relativity and test of independence
b Goodness of fit- does frequencies differ from distribution test of independence- are two categories of variables independent of eachother+ what is degree of association
I decide to use numerous t-tests for post hocing my anova... what is the term that describes the increase in error exponentially due to this problematic method a. standard error b. experimentwise error rate c. error of sample means d. type one fault
b this is unacceptable, we need a better approach so we shift to ANOVAS
I have a sample of men and women who do and dont exercise question: is gender associated with exercise? drawing a two by two I fill in my observed values. ME-20 FE-60 MDE-40 FDE- 80 total males-60 total females- 140 exercisers- 80 non exercisers-120 how would I get my expected values for exercisers? a. 80/400 to get .2 b. 80/200 to get .4 c. 120/200 to get .6 d. 120/400 to get .3
b we are dividing number of exercisers by total population to determine EXPECTED proportion of exercisers
which F value leads to a greater difference between groups means relative to variability within the groups? a. 1.11 b. 17.71 c. .005 d. -2.1
b (larger the value)
the F is significant in my ANOVA and I want to run a multiple comparisons test (post hock) which involves a rank ordering of means followed by successive contrasts of pairs of means what value am I likely going to be using to determine difference between largest and smallest mean values? a. p value b. f statistic c. q statistic d. alpha
c
the researcher tests the null hypothesis and determines that there is no difference between the exercising in water versus on land in patients with alopecia... in this case they would a. establish that null of the hypothesis is true b. reject null hypothesis c. fail to reject null hypothesis d. question their lives for asking such a dumb research question
c
two means differ a. null b. one sided alternative hypothesis (directional) c. two sided alternative hypothesis
c
unlike a z test, t tests uses what kind of sample variance in the denominator? a. normal b. optimal c. pooled/weighted d. standard
c
used with population based data What kind of test: a. t test b. anova c. z test d. chi squared
c
what are the only legitimate conclusions when looking at a null a. true or false b. proven or not proven c. reject or not reject d. accept or deny
c
what is the distribution of an anova? a. z score b. t score c. f value d. p value
c
what is the key thing about the scheffe test? a. most static against type II error b. most flexible and rigorous against type II error c. most flexible and rigorous against type I error d. it is type I error
c
what is the null of a t test? a. mean 1 is greater than mean 2 b. mean 1 is less than mean 2 c. mean 1 is equal to mean 2 d. a&b
c
which is not an assumption of a t test? a. IV is categorical/nominal and has two levels b. distribution of DV is normal c. distribution of IV is normal d. variances of the DV are similar (homogeneity)
c
which post hoc analysis: a. turkeys honestly significant difference b. newman-keuls method c. sheffe test d. bonferroni correction I wanted to look at the types of music in a clinic and the effect it would have on participation. I looked at rap, country, folk, pop, r&b, jazz, and reggae. In order to keep strong protection against type I error, I changed my critical value based on the number of comparisons being done
c
I am looking to see the means of proportions in a nonparametric test in which my numerator is the difference between sample means and my denominator is the SE of sample differences. If the z is greater than 1.96 I would reject the null and find an exact associated p value to judge whether or not proportions are equal. a. z test b. t test c. binomial d. chi squared
c (key words: nonparametric and proportion data)
which of these tests are typically known as parametric tests - logarithmic data -z test -t test - chi squared -anovas -binomial
z test, t test, anovas
We are looking to compare the outcomes of older adults with balance disorders treated in 3 clinics each of which plays different music in the background clinic 1- no music clinic 2- classical music clinic 3- rap music if you were to reject the null through an ANOVA... what would It tell you a.all groups are the same b. one of the groups are not equal c. all groups are not equal d. there is a difference, but it is not apparent where until you do a post hoc
d
When p is computed.... it us compared to present alpha (significance level).... the null is rejected if p is a. above alpha (significance level) b. at or above alpha (significance level) c. below alpha (significance level) d. at or below alpha (significance level)
d
You see an ANOVA in which the groups overlap quite a bit. Distance between each group is less than the distance within each group. a. statistically significant ANOVA, large F ratio b. statistically significant ANOVA, small F ratio c. not statistically significant anova , large F ratio d. not statistically significant anova , small F ratio
d
alli knows that she needs to set a significance level to have any relevance in her hypothesis test... when does she do this a. before hypothesis b. during test c. post test d. a priori
d
nonparametric tests are typically used with a. IR data b. ordinal data c. nominal data d. b&c
d
the F is significant in my ANOVA and I want to run a multiple comparisons test (post hoc) which involves a rank ordering of means followed by successive contrasts of pairs of means in this case what would be compared against a critical value a. single means b. group means c. single differences d. pairwise differences
d
the inferential steps to conclude that a null hypothesis may be false goes as follows: The data (or data more extreme) are very unlikely given the null hypothesis is true. this means that: a. a very unlikely event occurred b. the null hypothesis is not true c. the null hypothesis is true d. a&b e a&c
d
they, idk who they are, tell us that statistical tests are an expeditious approach to answering a singular questions... which is? a. do these two things occur together? b. is there a difference between samples or not? c. does this thing work for sure? d. is there a difference between populations or not?
d
what is the grand mean a. mean for one group in anova b. sum of squares of one group in anova c. sum of squares of within groups in anova d mean of the total sample in anova
d
what test is used with nominal data What kind of test: a. t test b. anova c. z test d. chi squared
d
which post hoc analysis: a. turkeys honestly significant difference b. newman-keuls method c. sheffe test d. bonferroni correction I wanted to look at the types of music in a clinic and the effect it would have on participation. I looked at rap, country, and folk. since I know that the number of comparisons is not too large.... I set the alpha rate and divide by the number of comparisons to keep and overall type I error rate of .05.
d
which reflects total variability in data a. SS within group-SS between groups b. SS within group+SS between groups c. SS total variation d. c&b
d
what contradicts the null, can indicate a direction of the difference of relationships you expect? a. alternative hypothesis b. null hypothesis c. research hypothesis d. a&c
d (alternative and research hypothesis are synonyms)
which of these tests are typically known as nonparametric tests - logarithmic data -z test -t test - chi squared -anovas -binomial
logarithmic data, chi squared, binomial
I do an experiment between three groups of athletes and determine the is some difference through the ANOVA, but am bummed because I couldn't tell ya exactly where that difference is. what would I have to do to find where difference lies?
post hoc analysis
binomial tests and chi squared tests are examples of?
nonparametric tests
I am looking at proportions of a sample that looks at wages (30k-50k, 50k-70k, 70k-110k) and career stage (early career and late career) for mu chi square... do these meet the assumptions or is there a need for continuity corrections. Reminder! ASSUMPTIONS FOR CHI SQUARE - frequency data - adequate sample size - mutually exclusive categories - solid theory for categorization of variables
test it does meet the needs
first everyone believed there was a wolf, when there wasnt what kind of error
type 1
next they believed there was no wolf, when there was what kind of error
type 2
We are looking to compare the outcomes of older adults with balance disorders treated in 3 clinics each of which plays different music in the background clinic 1- no music clinic 2- classical music clinic 3- rap music What is I called if I am looking at how much difference occurs just in the no music group
within group variance