STATS final test
Exercise: B = -14.29, sig = 0.02 If engaging in 2 hours of exercise, what is the predicted average change in the number of minutes of ritualistic behaviors when controlling psychological distress and number of counseling sessions? A. Decrease by about 29 minutes B. Increase by about 29 minutes C. No change in ritualistic behaviors D. Both decrease and increase are possible
A. Decrease by about 29 minutes
You are trying to determine if there is a significant difference in the mean Glasgow coma scale score for patients who are seen with a head injury during the day, evening and night shifts in your emergency room. The ANOVA results show an F value of 22.33 and a p-value <0.01 (alpha=0.01). This means: A. There is a significant difference in the mean Glasgow coma scale score for patients seen with a head injury on the three different shifts. B. There is NOT significant difference in the mean Glasgow coma scale score for patients seen with a head injury on the three different shifts. C. A type I error was made. D. A type II error was made.
A. There is a significant difference in the mean Glasgow coma scale score for patients seen with a head injury on the three different shifts.
A nurse researcher studies the impact of medication and exercise on HgA1C levels and reports that more exercise significantly decreases HgA1C levels for the same medication. This means: A. the beta coefficient associated with exercise is negative B. the beta coefficient associated with exercise is positive C. the coefficient of medication is higher than that of exercise D. the R square change associated with exercise is not significant
A. the beta coefficient associated with exercise is negative
A nurse lead study examines the relationship between age and post-operative pain. The Spearman's rho is -0.2 and the relationship is significant. This means: A. As age increases, post-operative pain decreases. B. As age increases, post-operative pain increases. C. There is a strong relationship between age and post-operative pain. D. Post operative pain is significant in the elderly.
A. As age increases, post-operative pain decreases.
A study examined the relationship between having a baccalaureate degree and passing a cultural competency exam among a group of 987 randomly selected registered nurses at your hospital. The researchers report that more registered nurses with a baccalaureate degree passed the cultural competency exam (OR 1.54, 95% CI 0.98-1.79). Interpret this information. A. More nurses with a baccalaureate degree is estimated to pass the exam but this was not a significant difference. B. Nurses with a baccalaureate degree were significantly more likely to pass the exam. C. 98% of the nurses passed the exam and the 2% that didn't did not have a baccalaureate degree. D. This is an example of a type II error.
A. More nurses with a baccalaureate degree is estimated to pass the exam but this was not a significant difference. (because the CI lower bound was <1)
A study utilizes repeated measures ANOVA to examine the impact of using 3 months of suppressive treatment with acyclovir immediately followed by 3 months of a new antiviral medication on HSV 2 lesions. The researchers conclude that the new antiviral medication is significantly better at reducing HSV 2 lesions. You are not confident in this results because you are concerned about: A. carry-over effects B. inadequate power C. homogeneity of variance D. inadequate sample size
A. carry-over effects
A study that examines the relationship between eating citrus fruit and getting cold sores finds a relative risk of 1.89 with a p value of 0.047. The results indicate at significance level of 0.05: A. individuals who eat citrus fruit are significantly more likely to develop cold sores B. there is not a significant relationship between eating citrus fruit and developing cold sores C. eating citrus fruit is a significant protective factor for developing cold sores D. if individuals have a cold sore and eat citrus fruit they can get better significantly faster
A. individuals who eat citrus fruit are significantly more likely to develop cold sores
Which of the following is true about the Hosmer-Lemeshow test in logistic regression? A. The null hypothesis is that the model doesn't fit the data. B. If the p-value for this test is less than .05, we conclude that the model does not fit the data. C. This test tells us whether at least one predictor is significantly associated with the outcome. D. The test detects the specific significant predictor(s).
B. If the p-value for this test is less than .05, we conclude that the model does not fit the data. goodness of fit test binary response small p value = poor fit
600 people are recruited in a study about how average weight loss (measured in pounds) is affected by exercise type and diet type. The researchers investigate 3 types of exercises (cardio, HIIT, and weight training) and 2 types of diet (low-fat and high-fat). The 600 people are evenly divided into 6 different groups so each group can receive a unique combination of one diet type and one exercise type. What kind of ANOVA would be used after the data are collected? A. One-way ANOVA B. Two-way ANOVA C. Repeated measures ANOVA D. Mixed ANOVA
B. Two-way ANOVA two independent variables (exercise and diet type)
A nurse researcher studies the impact of medication and exercise on HgA1C levels and reports that the coefficients of medication and exercise are both negative. The R associated with the model is 0.8 with p-value of 0.002. This means: A. medication and exercise are both significant predictors B. medication and exercise explain 80% of the variance in HgA1C levels C. medication and exercise explain 64% of the variance in HgA1C levels D. medication and exercise explain 20% of the variance in HgA1C levels
C. medication and exercise explain 64% of the variance in HgA1C levels
A researcher wants to determine how college GPA and engagement score (0-100) impact performance (high/low) for nurses enrolled in an online graduate programs. What test would you recommend she to utilize? A. ANOVA B. T-test for independent groups C. Logistic regression D. Pearson's Correlation Coefficient
C. Logistic regression LR analyzes a relationship between one or more predictions and a binary outcome. - predictors - categorical and numeric - response - logit(p) = B0 + B1+B2+B3.... or =a+b1x1+b2x2+..
You are trying to determine if there is a significant difference in the mean Glasgow coma scale score for patients who are seen with a head injury during the day, evening and night shifts in your emergency room. The ANOVA results (F statistic of 22.33 and p-value <0.01) show a significant difference in the mean Glasgow coma scale score for patients seen with a head injury on the three different shifts. How can we determine which shift(s) is different from other(s) in the mean Glasgow coma scale score? A. The ANOVA results already show all three shifts are different from each other. B. No test can specify the different group. C. Post-hoc t-tests D. An additional chi-square test.
C. Post-hoc t-tests
A psychiatric nurse practitioner completes a study examining psychological distress scores, hours spent sleeping, and number of visitors to see if these variables impact minutes spent in ritualistic behavior among patients diagnosed with obsessive compulsive disorder. She reports the following information. What statement is true? psychological distress B= 11.511, sig = 0.000 sleeping B=-4.735, sig = 0.02 visitors B=3.22, sig = 0.41 A. The most significant predictor is the number of visitors. B. All of the variables are significant predictors. C. Sleeping significantly impacts the number of minutes spent in ritualistic behavior. D. For each additional hour of sleep, ritualistic behaviors increases by 0.02 minutes if controlling the other two covariates.
C. Sleeping significantly impacts the number of minutes spent in ritualistic behavior.
A study utilizes ANOVA to examine the average distribution rates of an antibiotic in infants, toddlers and school age children. The results include an F statistic that is 1.04. You know this means: A. There is a statistically significant difference in the average distribution rates in the three groups. B. Infants distribute the antibiotic faster because they have a higher percentage of fluid in their bodies. C. The variation between the groups are similar to the variation within the groups. D. There NO statistically significant difference between the groups.
C. The variation between the groups are similar to the variation within the groups.
A researcher wants to examine the relationship between age in years and the engagement score (0-100) for 700 nurses enrolled in an online degree program. What test would you recommend? A. ANOVA B. T-test for independent groups C. Logistic regression D. Pearson's Correlation Coefficient
D Pearson's Correlation Coefficient is a statistic that measures linear correlation between two variables X and Y. It has a value between +1 and −1. A value of +1 is total positive linear correlation, 0 is no linear correlation, and −1 is total negative linear correlation.
You randomly select 120 healthy individuals and randomize them into a group given a daily multivitamin and a group who does not take a daily multivitamin. You follow the sample for ten years and identify all cases of pernicious anemia which develop. Which statement is true? A. You can only determine the odds ratio for pernicious anemia B. You should NOT determine a relative risk since this is a cross sectional design C. Your exposure of interest is retrospective D. You can determine the RR for developing pernicious anemia from this data
D. You can determine the RR for developing pernicious anemia from this data (because this is a cohort study, if is was cross sectional it would be Chi^2)
what are 5 nonparametric tests and their parametric counterpart?
Mann-Whitney U test, aka Wilcoxon rank-sum test (alternative to two sample t-test) Wilcoxon signed-rank test (alternative to paired t-test) Kruskal-Wallis test (alternative to ANOVA) Friedman test (alternative to repeated measure ANOVA) Chi-square test
Which test would be most appropriate to answer the following research question: Do patients with a history of cocaine use have more gumline cavities than other patients? There were 15 people in the sample (7 with a history of cocaine use and 8 without), and the number of gumline cavities was not normally distributed. Independent t test Mann-Whitney U-test Paired t test Wilcoxon signed-rank test
Mann-Whitney U-test
chi^2 vs mcnair test?
McNemar tests as a repeated measures version of a chi-square test of independence. Chi-Square tests can be used for larger tables, McNemar tests can only be used for a 2×2 table. Chi-Square is testing for independence between two nominal or ordinal variables organized in a bivariate table, nonparametric test. McNemar is testing for consistency of responses across two variables.
Researchers studied symptom distress and palliative care designation among a sample of 710 hospitalized patients. Controlling for age, they used a t-test to compare average distress from nausea scores in men and women. Lower scores indicated less distress from nausea. They report men had an average score of 1.02 and woman had an average score of 1.79. Which statement is correct?
Men had less distress from nausea on average than women but we can not determine if this is a significant difference.
Your study of the use of compression stockings for post-surgical DVT prophylaxis utilizes an independent t-test and reports a p value of 0.02. The study has an alpha of 0.05, a beta of 0.08 and a sample size of 567 patients. What do you conclude? A. The results are clinically significant B. The results are not statistically significant C. The results are statistically significant. D. A type II error is being made
The results are statistically significant.
Researchers study the relationship between interpersonal violence and health in college age women. The selected an alpha of 0.05. The researchers examined levels of social support for abused versus non abused women. A p value of 0.19 is reported. Based on this information, you know:
There is no significant difference in social support between the two groups.
A small study reports energy drinks are associated with hyperadrenergic syndrome with an effect size of 0.8. You are funded to complete a larger study on the issue. You know:
This is a strong effect size so you will probably only need a small sample to detect the association.
2 sample t test
aka independent t test: "The equality of the means of two independent groups" compare means of 2 populations: with 2 mean values, variances and sample sizes. Then calculate the p value. alternate is MANN-WHITNEY U TEST, AKA WILCOXON RANK-SUM TEST
You conduct a case control study and determine that 5 people with Guinea Worm were exposed to water from river A. One person with the disease had no exposure to water from river A. You also determine that ten people without the disease drank from river A and eighty-nine were healthy and did not drink from river A. Calculate the appropriate Odds ratio. A. 14.3 B. 44.5 C. 15 D. 5
b 44.5 (5/15) = 0.33/1-0.33 = 0.499 (1/90) = 0.011/1-0.011 = 0.011 0.499/0.011 = 45.4 OR AD/BC = 5*90/10/1
repeated measure ANOVA
testing The equality of the means of multiple *DEPENDENT* groups. used when 2+(usually 3+) data sets being compared and repeated measures accounts for data must test for sphericity alternate is Friedman test, (Test any difference in medians of two or more related groups. Friedman test is an analysis of variance of the ranks)
what tests should be used when limitations to the chi squared test are apparent?
when the data has a low frequency use fisher exact test when the data is paired use McNemar test
28 healthy adults who consumed a study-prepared high-aspartame diet (25 mg/kg body weight/day) for 8 days and a low-aspartame diet (10 mg/kg body weight/day) for 8 days, with a 2-week washout between the diets, were examined to determine the neurobehavioral effects of consuming both low- and high-aspartame diets. For each adult, the working memory score (0-1000) was recorded twice (once after high-aspartame diet period and once after low-aspartame diet period). Suppose all assumptions are satisfied, which of the followings is most appropriate for the difference in the mean working memory scores between the two diets? - two-samples t test - one sample t test - Brown-Forsythe test - paired t test
- paired t test
If a study has an alpha of 0.05 and a power of 0.80 you know the chance of making a type II error is:
20% but i don't know how to calculate
A study examining the relationship between the number of quiet hours and patient satisfaction score reports r=0.6, p=0.02. You know this means: A. 36% of the variance in patient satisfaction can be explained by different numbers of quiet hours. B. There is a weak relationship between the number of quiet hours and patient satisfaction. C. There is a significant relationship between the number of quiet hours and patient satisfaction because the p value is less than the r value. D. For each additional quiet hour patient satisfaction increases 60%.
A. 36% of the variance in patient satisfaction can be explained by different numbers of quiet hours.
There is a study on gastroesophageal reflux therapy among two hundred premature infants all of whom are less than 30 weeks and being tube fed in a NICU. You measure the average intraesophageal pH upon entry into the study, at two weeks, at four weeks and at six weeks. The best analysis method is: A. repeated measures ANOVA B. Pearson's correlation coefficient C. regression using a model that includes time, volume of feeds and intraespohageal pH D. Chi-square
A. repeated measures ANOVA the variables are dependent upon each other, ie same baby through time.
A study compares the average blood sugar in young adults, middle aged adults and the elderly after receiving a Thiazolidinedione drug. An appropriate test would be: A. Chi-square B. ANOVA C. dependent t-test D. independent t-test
ANOVA
what is the significance if the f statistic is close to 1? what test is this
ANOVA If the F statistic is close to 1, the variation between groups is close to that within groups so the result is unlikely significant. But always determine significance using p value!
One-way ANOVA
Analysis of variance. Testing the equality of the means of multiple *INDEPENDENT* groups. A statistical test used to analyze data from an experimental design with one independent variable that has 2 or more groups (levels). (usually 3 or more) alternate is KRUSKAL-WALLIS TEST (two or more?)
A psychiatric nurse practitioner completes a study examining psychological distress scores (DS), hours spent exercising (X), and number of counseling sessions (C) to see if these variables impact minutes spent in ritualistic behavior among patients diagnosed with obsessive compulsive disorder. She reports the following regression equation.Y= 59+ 3.98(DS)-2.89(X)-4.5(C)If a patient has a distress score of 5, spends 2 hours exercising, and attends a counseling session, you would expect how many minutes of ritualistic behavior? A. 9.6 B. 68.6 C. 29.5 D. unable to determine
B. 68.6
A study examines the relationship between total score on a cultural competency exam among a group of 987 randomly selected nurses at your hospital and their years of educational preparation, language spoken, age, primary nursing role, race and ethnicity. What test would be appropriate to determine the association? A. Independent t-test B. Multiple linear regression C. ANOVA D. Pearson's correlation
B. Multiple linear regression
A logistic regression model log it(p) = B0+B1(age)+ B2(gender) was run looking at the association between the response diabetes (yes = 1, no = 0), and predictors age and gender (male = 0, female = 1). The e^B1=2.5 with a 95% CI [1.8, 3.3]. This means that A. The association between age and diabetes is not statistically significant (i.e. p>0.05). B. We are 95% confident that the odds ratio of people to those one year younger in diabetes is between 1.8 and 3.3 C. We are 95% confident that the odds ratio of women to men for diabetes is between 1.8 and 3.3 D. Increased age is associated with lower odds of diabete
B. We are 95% confident that the odds ratio of people to those one year younger in diabetes is between 1.8 and 3.3
A study examines the risk of overdose (yes/no) and the use of hydromorphone (yes/no) in a sample of independent teens. What would be an appropriate test to use? A. Dependent t-test B. Chi-square C. Independent t-test D. McNemars test
B. Chi-square the outcomes are nominal, and the distribution is not known?
Main tests for equality of variances
Brown-Forsythe test (most powerful) Levene's (mean) test Test the equality of variances using the sample! (Equal variances assumption is paramount, so graph analysis is not reliable)
A logistic regression model log it(p) = B0+B1(age)+ B2(gender) was run looking at the association between the response diabetes (yes = 1, no = 0), and predictors age and gender (male = 0, female = 1). The Exp(B2)=3.4 with a 95% CI [2.9,4.6]. This means that A. The association between gender and diabetes is not statistically significant (p>0.05). B. The odds of diabetes for men is 3.4 times that of women. C. The odds of diabetes for women is 3.4 times that of men. D. The chance of diabetes for women is 3.4 times that of men.
C.The odds of diabetes for women is 3.4 times that of men. Exp(B) - This is the exponentiation of the B coefficient, which is an odds ratio. log(odds) the odds ratio would be <1 if it favored men because men = 0 and women = 1 in this equation?
what is the difference between relative risk and odds ratio?
Cohort study usually uses relative risk Case control and cross-sectional study use odds ratio rr= (a/(a+b))/(c/(c+d)) OR = ad/bc
Several days after a wedding, an outbreak of cyclosporiasis occurred among attendees. Of the 108 guests at the wedding, 76 were ill and met the case definition. 50 of the 76 who were ill ate the wedding cake. Three of those who were well ate the wedding cake. What is the attack rate for those who ate the wedding cake? A. 46.3% B. 65.8% C. 70.3% D. 94.9%
D 94.9% 50/(50+3) = .9434 attack rate: is the incidence proportion of exposed people. #new cases in exposed group/exposed population.
A study examines the use of synthetic opioids and neuropathic pain. The Chi-square value is 2.71. You know this means: A. There is a statistically significant association between the use of synthetic opioids and neuropathic pain. B. There is no association between the use of synthetic opioids and neuropathic pain. C. There is a clinically significant association. D. Unable to determine.
D Unable to determine. need a p value an alpha? or df
Which of the following is the nonparametric analog of a repeated measures ANOVA? Wilcoxon rank sum test Friedman test Wilcoxon signed rank test Kruskal-Wallis test
Friedman test
Interpreting R2
Interpretation: r^2 measures the proportion of the variation in y(DV) that is explained by x(IV). sensitive to outliers
What is Pearson's r?
It's a measure of the strength and direction of the linear association between two variables. ρ(rho) is the population correlation r is the sample correlation. +1 is perfect positive correlation -1 is perfect negative correlation 0 is a weak correlation.
Among entering kindergarten students, does a relationship exist between the number of letters of the alphabet recognized and the type of dwelling in which the child lived (i.e., house, apartment, and mobile home)? The sample contains 33 children, and the data are NOT normally distributed. Which of the following tests is the best for the study? Mann-Whitney U-test Wilcoxon signed-rank test two-samples t test Kruskal-Wallis test
Kruskal-Wallis test: letters in name and dwelling(3+ IVs) are 2 independent groups (not paired) Alternative to ANOVA
what is the difference between a paired t-test and a 2 sample t-test?
Paired-samples t tests compare scores on two different variables but for the same group of cases; independent-samples t tests compare scores on the same variable but for two different groups of cases.
Describe the difference of SLR, MLR, and LR
SLR - a relationship between 2 quant variables in a straight line. only one IV (X), predict value of Y for any X MLR - similar to SLR but has multiple IVs LR - relationship between one or more predictors and a binary outcome.
Which regression model is most appropriate for the following scenario? The data of 55 people were used to analyze the relationship between the amount (in min) of exercise per week and resting heart rate?
SLR?
Which of the followings is the best test (most powerful) for normality? Lilliefors corrected K-S test Shapiro-Wilk test Kolmogorov-Smirnov (K-S) test Brown-Forsythe test
Shapiro-Wilk test
A professor wishes to determine if there is a difference between the average test grades of her students who use the resources available to them with those students who don't use the resources. The alpha is 0.05 and the P-value is 0.03. She should conclude:
She can reject the null hypothesis and there is a statistically significant relationship between nursing students' grades and their use of available resources.
what is spearman's ρ?
Spearman's rank correlation coefficient, aka Spearman's ρ, is a numerical measurement of the strength and direction of the monotonic relationship between two **ordinal or **numeric** variables nonpara of pearson's r. - do not need normality, resistant to outliers d/t ranks not original data, measures monotonic instead of linear relationship.
How to verify the normality so we can perform parametric tests? (3) what if the p-value is > alpha?
Test of normality: - Shapiro-Wilk test (most powerful) - Lilliefors corrected K-S test - Kolmogorov-Smirnov (K-S) test (least powerful of the 3) p>alpha fail to reject null (use parametric, assume normality) p<alpha reject null
28 healthy adults who consumed a study-prepared high-aspartame diet (25 mg/kg body weight/day) for 8 days and a low-aspartame diet (10 mg/kg body weight/day) for 8 days, with a 2-week washout between the diets, were examined to determine the neurobehavioral effects of consuming both low- and high-aspartame diets. For each adult, the working memory score (0-1000) was recorded twice (once after high-aspartame diet period and once after low-aspartame diet period). The t test result shows that there is no significant difference (p-value>0.05) in the mean working memory scores between the two diets. Without any calculation, what can we say about the 95% confidence interval of the mean difference? The 95% confidence interval does NOT include 0. The 95% confidence interval includes 0. The 95% confidence interval doesn't exist.
The 95% confidence interval includes 0.
Pearson's r
The correlation between two numeric variables (-1 to +1) Spearman's ρ is the nonparametric counterpart
In a t-test situation, magnitude of effects (clinical significance) is often communicated through: The absolute value of the t statistic The p value The magnitude of Levene's F The magnitude of Cohen's d
The magnitude of Cohen's D
The researcher used Wilcoxon signed-rank test to compare the number of times in a month people exercise before and after a 3-hour exercise orientation program. The researcher obtained a p value of 0.03. How should the researcher conclude the result?
The researcher should reject the null hypothesis. There is sufficient evidence to conclude an association between taking the 3-hour exercise orientation program and mean number of times people exercise.
You are reading a pilot study that reports taking glargine insulin prevents glycemic spikes. The p value is 0.08. You know this means:
The selected alpha was greater than 0.08.
A small pilot study reports consumption of energy drinks are not associated with adverse outcomes in children. However, a larger follow up study correctly reports that energy drinks are associated with death in children under age 5. You know this means:
The small study made a type II error.
When should Kruskal-Wallis test be used instead of Mann-Whitney U test? There are more than two groups to compare The data are not normally distributed The groups are related They are always interchangeable
There are more than two groups to compare Kruskal-Wallis test is alternative to ANOVA (2+ groups independent variables) the mann test is testing independent groups but only 2.
what is a fisher exact test?
Use when at least one cell in a 2x2 table is <5 (too small for Chi-square). ? add 1 to the right column?
Which of the following statistical tests can be used if the data are paired but not normally distributed? Two-samples t test assuming equal variances Paired t test Wilcoxon signed rank test Mann-Whitney U-test
Wilcoxon signed rank test
Patients in a critical care unit, who are matched on age, receive either experimental oral care or control oral care. Which test is the best to determine whether the type of oral care have an effect on the length of patient's stay in the critical care unit? There are 20 pairs of patients, and their length of stay was not normally distributed. Mann-Whitney U-test Paired t test Wilcoxon signed-rank test
Wilcoxon signed-rank test alternative to paired t test
A researcher utilizes a t-test to compare the average amount of beta-lactamase produced by two different types of bacteria. The Levene's test for equality of variances has a p value of 0.09. You know this means: A. You can NOT report the t-test results associated with assuming equal variances. B. The t-test is statistically significant so you reject the null hypothesis. C. The t-test is not significant so you fail to reject the null hypothesis. D. You can report the t-test results associated with assuming equal variances.
You can report the t-test results associated with assuming equal variances. - Levene's test is used to check that variances are equal for all samples when your data comes from a non normal distribution. - If the p-value is larger than the alpha level, then you can say that the null hypothesis stands — that the variances are equal; if the p-value is smaller than the alpha level, then the implication is that the variances are unequal.
Your patient completes a DXA scan and the report indicates, compared to the mean bone density of normal population, her left hip bone density has a Z score of -1.4 with a p value of 0.001 (alpha is set at 0.01). You know this means:
Your patient's left hip bone density is significantly below average.
a study has an inadequate sample size you increase the risk of: A. a small effect size B. a type I error C. a type II error D. a large effect size
a type II error
If you are very concerned about making a type I error what would be the best alpha to select?
a very small alpha ie. 0.01
paired t test
aka dependent t test: "testing The equality of the means of two independent groups" a t-test performed on a repeated measures two-group design Test the equality of the means of two dependent groups Data either: The same individuals tested on two occasions or under two different conditions Two groups of subjects that are matched or paired on one or more characteristics (e.g., IQ, age, gender, etc.) alternate is WILCOXON SIGNED-RANK TEST
A pilot study examines the concurrent use of dextromethorphan and SSRIs (yes/no) and serotonin syndrome (yes/no). The alpha was 0.10 and the power was 0.80. The p value was 0.03. You know this means: A. Use of dextromethorphan increases the risk of serotonin syndrome. B. There is no association between concurrent use of dextromethorphan and SSRIs and serotonin syndrome C. There is a clinically significant association between concurrent use of dextromethorphan and SSRIs and serotonin syndrome D. There is a statistically significant association between concurrent use of dextromethorphan and SSRIs and serotonin syndrome.
d. There is a statistically significant association between concurrent use of dextromethorphan and SSRIs and serotonin syndrome.
If the null hypothesis mean=0, the power of a test is higher when true mean=2 compared to a true mean=3. tf
false
Describe how to find the linear regression line of best fit to data using your own words.
finding a line that is the smallest distance from each of the data points on average.
attack rate formula
ill/(ill+well) x 100 this is in the exposed group
Odds Ratio (OR)
in a 2x2 table ad/bc Case control and cross-sectional study use odds ratio
A researcher wishes to determine if nurses in California have higher average vitamin D levels than nurses in New York. What would be an appropriate test to use? A. Independent samples t test B. Chi-Square C. Paired t test D. Pearson's Correlation Coefficient
independent samples t test (aka 2-sample t-test) The Independent Samples t Test compares the means of two independent groups in order to determine whether there is statistical evidence that the associated population means are significantly different.
A health department investigation determines the following after an outbreak: What exposure was a significant risk factor? select all that apply. exposure, OR, 95%CI Meatballs 4.5, 1.25-6.74 Ice cream, 2, 0.98-4.71 salad, 0.78, 0.5-1.78 soup, 1.03, 0.23-2.94
just the meatballs this is the only one where the low end of the CI is > 1. this means that we are 95% confident that the meatballs are a significant risk factor
Which regression model is most appropriate for the following scenario? A study of 114 college students is performed to study the relationship between a college student's residence (lives on-campus vs. off-campus) and current smoking behavior (smokes or does not) adjusting for age?
logistical regression?
what is the relationship with power and; sample size, standard deviation, effect size?
power goes up if: ↑sample size (more people more powa) ↓standard deviation ↑effect size
what is the table for statistical and clinical significance? small and large sample....
sample size, P-value, statistically sig, clinically sig. - Large, <alpha, yes, ? - Small, <alpha, yes, probably yes - Small, >alpha, no, ? - Large, >alpha, no, no
The Kruskal-Wallis test would be better than a one-way ANOVA when the independent variable has more than two categories. the dependent variable is not normally distributed you want to know whether there is a difference in means between two groups. the sample size within each group is large
the dependent variable is not normally distributed
What is effect size?
the size of the difference between group means that exists within the population. Effect size and sample size are inversely proportional: As one increases, the other decreases. the bigger the difference the smaller sample size needed to see the difference. A weak effect size < is 0.3 (or -0.3). A moderate effect size is 0.3-0.5 (or -0.3 to -0.5). A strong effect size is > 0.5 (or -0.5).
The Wilcoxon signed rank test is best used when: there are two matched groups to compare there are three or more matched groups to compare the data are normally distributed the data are from an entire population rather than just a sample
there are two matched groups to compare
the required sample size increases as effect size decreases. tf
true
Which of the followings is NOT a nonparametric test? two-samples t test Mann-Whitney U test Signed rank test Kruskal-Wallis test
two-samples t test
If a study reports INCORRECTLY that an association exists this can only be what type of error?
type 1
A study reports that the abuse of bath salts (synthetic cathinones) is associated with developing candida. If this result is an error you know this means:
type 1 error was made