Biological Statistics
Binomial Distribution (information needed)
# of trials Pr[success] # of successes interested in calculating a probability for (x) Pr[failure] (Q=1-p)
Calculate the odds: Over the course of 2 months, I recorded the number of times hummingbirds fed at hyssop plants. Hummingbirds fed at hyssop plants 12 times and flew by without feeding 7 times. What are the odds of hummingbirds feeding at hyssop?
(12/19)/(7/19) = 1.71
Calculating odds
(p/1-p). The p variable is the OUTCOME (probability it will occur/prob. it wont) *flip formula for odds against*.
Approximation of Confidence Interval
+/- 2x standard error
Normal skew
0 Null: skew = 0
Which of the following significance levels can be used for calculating a confidence interval for a normal distribution?
0.01 0.10 0.20 0.05
You sample 20 individuals to see if they have the MC1R gene on chromosome 16. The gene is present in only 2 individuals. In what proportion of the sample was the MC1R gene present?
0.1
The probability of a randomly selected observational falling within the mean plus-or-minus 1 standard deviation (m+/- s) is ___________ in a normal probability distribution.
0.683
The standard normal distribution has a mean = _______ and standard deviation = ______.
0; 1
Under the alternate hypothesis, the F-ratio is expected to be much greater than ______.
1
Under the alternate hypothesis, the F-ratio is expected to be much greater than______.
1
Calculating standard deviation
1. Calculate each score's deviation (distance form the mean) 2. Square each deviation 3. Compute the mean for the squared deviations (this is the variance) 4. Take the square root of the variance (this is the standard deviation)
The methods section of a journal article states that the researchers conducted a 3x4 factorial analysis of variance with 10 replicates of each treatment. What is the total sample size for this experiment?
120
The methods section of a journal article states that the researchers tested 4 treatments using a randomized complete block experimental design and analyzed using ANOVA methods. They also state that they had 5 blocks. What is the total sample size for this experiment?
20
Normal kurtosis
3 Excess kurtosis = 0 (null)
Which of the following fail-safe numbers would lead you to feel most confident in the conclusions of a meta-analysis that included 30 studies?
32
You sample 20 individuals to see if they have the MC1R gene on chromosome 16. The gene is present in only 1 individual. In what percentage of the sample was the MC1R gene present?
5%
1 standard deviation
68% of data
An agronomist wanted to know whether a new type of fertilizer would increase yields enough to justify paying its high cost. They applied their old fertilizer to five fields and the new fertilizer to five fields. At the end of the growing season, crop harvests in the fields treated with the old fertilizer were 50 bushels per acre yield on average. They conducted a t-test and found an t-statistic = 1.275, and a p-value =0.057. If variances were equal, how many degrees of freedom would this test have?
8
Which of the following are measures of uncertainty that can be calculated. From sample data?
95% confidence interval Standard error
2 standard deviations
95% of data
3 standard deviations
99% of data
Probability distributions
A listing of all possible outcomes, and the probability of each occurrence.
Multiplication rule
A rule of probability stating that the probability of two or more independent events occurring together can be determined by multiplying their individual probabilities. Probability of of AB and Rh- occurring together = 0.041 x 0.146 = 0.006
Inductive Reasoning
A type of logic in which generalizations are based on a large number of specific observations. Karyotype 100 people and find 46 chromosomes in all. Conclusion: all people have 46 chromosomes.
For the characteristic provided, identify if it applies to t-tests only, analysis of variance (ANOVA) tests only, both t-tests and analysis of variance tests, or neither t-tests nor ANOVA. Test is based on partitioning variance due to treatment effects from variance due to sampling error.
ANOVA only
A peer researcher can use the digital object identifier to download a full dataset of interest or at least the metadata that went along with the data.
Accessible
Which one of the following best identifies a characteristic of a t-distribution in terms of how it differs from the standard normal distribution (Z-distribution)?
Accounts for uncertainty due to sampling error
Which of the following best represent when the Tukey-Kramer adjustment is used for pairwise comparisons?
After an ANOVA results in a significant finding (ANOVA p<0.05)
Among brown and yellow Labrador retrievers (labs) bred in the U.S., a researcher hypothesizes that brown labs are more likely to be aggressive than yellow labs. Which one of the following most appropriately identifies the statistical populations related to this hypothesis?
All brown labs and all yellow labs in the U.S.
In a phase 1 study testing the efficacy of a new drug therapy for heart disease, researchers randomly select 1000 participants with the targeted type of heart disease. Of these, 500 are randomly assigned to the treatment group receiving a pill with the drug. The other 500 are given a placebo pill that does not contain the drug. What is the statistical. Population?
All individuals with the targeted heart disease
In a phase 1 study testing the efficacy of a new drug therapy for heart disease, researchers randomly select 1000 participants with the targeted type of heart disease. Of these, 500 are randomly assigned. To. The treatment group receiving a pill with the drug. The other 500 are given a placebo pill that does not contain the drug. What is the statistical population?
All individuals with the targeted heart disease.
Inaccurate, imprecise
All over the place
Which two of the following experimental designs could be implemented to account for additional variation that would otherwise contribute to the error term in an analysis of variance?
Analysis of Covariance Randomized Complete Block Design
Researchers wanted to know if an antibody treatment could boost immune system function against HIV. The antibody treatment was given to 25 mice who were then injected with HIV. Twenty-five control mice were also injected with HIV but had not been given the antibody treatment. Mice treated with the antibody had an average of 87% immune function, while control mice had 18% immune function on average. They conducted a t-test and found a t-statistic = 6.275 and a p-value = 0.0053. Which of the following is the most informative and appropriate biological conclusion for the researchers to draw based on this information?
Antibody treatment significantly increased immune function
You are evaluating normality of a dataset containing three different treatments (control, treatment 1, and treatment 2) with the response variable (gene length). You run tests for skewness, kurtosis, and the Shapiro-Will test using the stat.desc function in R. You are using alpha 0.05. How would you interpret a Shapiro-Will test result with W statistic = 0.960 and p-value = 0.394 in control treatment?
Based on p-value = 0.394, conclude gene length follows a normal distribution
For each consideration listed below, indicate the probability distribution to which it applies. Two and only two mutually exclusive outcomes are possible for each trial.
Binomial distribution only
Tests appropriate for 1 categorical variable with 2 mutually exclusive outcomes
Binomial test and x2 goodness of fit test
For the characteristic provided, identify if it applies to t-tests only, analysis of variance (ANOVA) tests only, both t-tests and analysis of variance tests, or neither t-tests nor ANOVA. Assumes sample units are independent of one another and drawn randomly from the statistical population.
Both
For each consideration listed below, indicate the probability distribution to which it applies. Distribution assumes the observed response variable is categorical or discrete numeric.
Both binomial and Poisson distributions
For each consideration listed below, indicate the probability distribution to which it applies. Probabilities sum to 1 over all mutually exclusive outcomes.
Both binomial and Poisson distributions
Describe an appropriate sample and sample unit for the previous question.
Brown and yellow labs sampled from each state in the US. Sample u nit would be one lab.
According to the [A], if a population has a normal distribution then it will also have a normally distributed sampling distribution.
Central limit theorem
Is the median of a dataset a measure of central tendency or spread?
Central tendency
Power
Chance of detecting a difference
Which of the following tests could be used to analyze the results of a study in which the relationship between birth month and handedness (left vs. right) was examined? The null hypothesis was that the frequency of right—handed individuals was independent of birth month.
Chi-square contingency analysis
Inaccurate, precise
Clustered close together away from the bulls eye
Decrease sample variance
Completely random design or randomized complete block design
You have a dataset form an experiment with one factor and 3 groups for that factor. You find that data in group 2 is highly non-normal and none of the transformations improve normality. How do you proceed?
Conduct a Kruskal-Wallis test using the raw data values.
Researchers wanted to know if dolphins in the Southern Hemisphere swim in a predominantly counterclockwise direction while sleeping. They recorded the percentage of time spent swimming clockwise during sleep for eight dolphins. Which of the following would you do first ask part of the analysis process?
Conduct a Shapiro-Wilk test
Addition rule
Considering mutually exclusive events, the probability of both occurring is the sum of the probabilities of each event. Probability of B or AB blood occurring= 0.122 + 0.041 = 0.163
In a study looking at a child's height (measured to the nearest 0.01 cm) at 2 years of age as a predictor of adult height (measured to the nearest 0.01 cm), height measurements are ________ data.
Continuous
Reducing Bias
Control Treatments, Random Treatment and Assignment, Blinding
Samples
Control and treatment groups
In a phase 1 study testing the efficacy of a new drug therapy for heart disease, researchers randomly select 1000 participants with the targeted type of heart disease. Of these, 500 are randomly assigned to the treatment group receiving a pill with the drug. The other500 are given a placebo pill that does not contain the drug. What is the sample?
Control group and treatment group
Which of the following quantities are used to calculate the standard error for Pearson's correlation coefficient (r)?
Correlation Coefficient Degrees of Freedom
Which of the following are used to calculate the test statistic for a correlation analysis?
Correlation coefficient (r) Standard error of r
Metadata
Data about data; methods and protocols followed in the collection of data, context, etc.
Nominal data
Data which consists of names, labels, or categories.
Limiting sampling error
Decrease sample variance due to extraneous factors, increase replication to capture more of the population, ensure equal replication for each treatment to provide balance
Researchers examined the frequency of accidents among vehicles of different colors with the null hypothesis that the frequency of accidents a given vehicle color was involved in was proportional to the number of vehicles of that color. They observed 234 vehicles of 8 different colors and recoded the number of accidents among those vehicles. A chi-square analysis resulted in a chi-square test statistic = 6.01
Degrees of Freedom= 7 Fail to reject the null
Which of the following quantities are used to calculate Pearson's correlation coefficient (r)?
Deviation of x Deviation of y Total sum of squares for y Total sum of squares for x
Deviation
Difference between an observation and the sample mean
Which of the following best describes the term 'residual' in regression?
Difference between the y-value of the linear regression line at the same x-value
In a study looking at the number of covid-19 patients in 5 different hospitals, the number of patients is ____________ data.
Discrete
When doing a systematic review or meta-analysis, which of the following is the last step that should be completed among those listed?
Draw conclusions form the body of evidence relevant to the specific question
Describe conclusions based on inductive reasoning
During a floristic inventory, a researcher noted that most cacti were found growing on clay soils and few grew on sand soils. They concluded that cacti grow best on clay soils. A researcher observed patients diaagnoseed with breast cancer also tended to have wet ear wax and concluded that breast cancer causes wet ear wax.
Chi-square distribution
Either 1 or 2 categorical variables (categorical or discrete) Mutually exclusive outcomes sum to 1
A researcher is studying the efficacy of a potential new sterilization drug for feral horse populations in the inter mountain western US. A random selection of 250 female feral horses is identified as meeting the study criteria, 125 of which are randomly assigned to receive the drug. The drug is administered by local wildlife officials to those 125 horses via a shot delivered by tranquilizer gun. The other 125 females are the control group and are not administered a shot. The number of foals born to each of the 250 horses are monitored for 2 years by researchers with no knowledge of which horses received the drug. What is a strategy to reduce bias?
Equivalent treatment of control group
Which of the following is/are an example of how QA/QC can be done before data collection?
Establish a standard degree of precision and units of measurement. Create clear protocols detailing how data are to be collected.
Sampling
Examining random individuals from a population then using that to describe the statistical population
Which of the following are assumptions of both the chi-square goodness-of-fit test and the chi-square contingency test?
Expected frequencies are five or greater in more than 80% of mutually exclusive outcomes. The outcome is mutually exclusive for each explanatory variable. The outcome of each trial is independent. No mutually exclusive outcome has an expected frequency less than one.
Which of the following are acceptable methods for assigning experimental units to treatment and control groups?
Experimental units are assigned a number, then numbers are written on same-sized pieces of paper and placed in a sock. The researcher blindly pulls papers form the sock to identify which experimental units are assigned to each treatment. Experimental units are assigned a number, then a random numbers generator in a statistical software program is used to identify which numbers should receive each treatment.
The test statistic for a regression analysis is ___________.
F-statistic
The test statistic for a regression analysis is:
F-statistics
Which of the following can be used to evaluate whether there is publication bias for a given topic?
Fail-safe number Funnel plot
Assumptions of general linear models are not the same as those for one-way ANOVA.
False
Correlation analysis can be used to analyze cause-effect relationships between explanatory and response variables
False
Sample size calculations can be performed for many different types of data using a single equation.
False
Sample size calculations can only be performed for designing experiments with alpha = 0.05 and desired power = 0.80.
False
The binomial test would be appropriate for testing the following null hypothesis Grizzly bears are randomly distributed throughout Yellowstone National Park
False
Which of the following are acceptable. Forms of data analysis
Feeding data from a study into an established mathematical model to predict a longer-term outcome Combining data from similar but separate studies to look for broad patterns and propose new hypotheses Calculating descriptive statistics, such as mean and standard deviation. Conducting inferential statistics to test hypotheses.
Data/metadata for a project are available in an online data repository and the digital object identifier is provided in the associated peer-reviewed publication
Findable
FAIR principles for data sharing
Findable, Accessible, Interoperable, Reusable
A meta-analysis project in which all studies evaluated are expected to share one same treatment effect with only sampling error contributing to differences in effect sizes between studies would be analyzed using which one of the following?
Fixed-effects model
Which of the following best represents the null hypothesis of the Shapiro-Wilk test?
For a given sample, data follow a normal distribution
Accurate, imprecise
Gathered around the bulls eye, but not closely
Which of the following are benefits of well-documented process workflows?
Helps research remember how data were manipulated. Statistical scripts are organized; analytical steps are described in detail. Allows research to focus efforts on testing specific hypotheses. Helps peer researchers reproduce results.
In meta-analysis, more weight is given to studies with _________ precision.
Higher
When doing a systematic review or meta-analysis, which of the following is the first step that should be completed among those listed?
Identify a specific question to be addressed
Shapiro-Wilk
If >0.05, normal and fail to reject null
A peer research can download a full dataset and be assured that the metadata is written using a standard vocabulary for that field and data files do not need a specific type of software to be opened
Interoperable
Which of the following are true of the concept of a significance level?
Is also called alpha Represents the probability at or below which we will reject the null hypothesis
Which of the following are associated with Type I error?
Is the same as the significance level Represents probability of rejecting the null hypothesis when it is true in reality
Which of the following are true of the concept of a significance level?
Is the same as type I error Represents the probability at or below which we reject the null hypothesis
Skew
Location of outcome with highest probability relative to other outcomes Positive: most values are on the lower end negative: most values are on the upper end 0: normal distribution
You conduct an experiment of randomly selected patients having ga range of low density lipoprotein (LDL) cholesterol concentrations. After five years, you determine which of those patients are still alive versus which have not survived to examine high LDL cholesterol increases the probability of mortality. What type of analysis would be most appropriate?
Logistic regression
Which of the following terms is most analogous to the calculation of variance?
Mean square
Which of the following are measures of central tendency for sample data?
Mean, median, mode
Which one of the following terms best describes research into the methods for, evaluation of, and incentives driving research practice?
Meta-research
In meta-analysis, factors that might help explain differences in effect size or outcomes between studies is called a ____________.
Moderator variable
For study conducted on patients from 5 different randomly selected hospitals, the name of the hospital is ____________ data.
Nominal
Which of the following theoretical probability distributions do not assume a sample units are randomly selected and independent when applied to statistical hypothesis testing?
None of the above
Which of the following theoretical probability distributions do not assume sample units are randomly selected and independent when applied to statistical hypothesis testing?
None of the above Binomial distribution Poisson distribution Normal distribution
Proportion
Number of observations in a given category/total observations
Which of the following terms is calculated as the number of groups minus one?
Numerator degrees of freedom
Discrete data
Numerical data values that can be COUNTED, whole numbers
Continuous Data
Numerical data values that can be MEASURED
Types of studies
Observational, experimentation, mathematical
Poisson Distribution Assumptions
Occurrences are random and independent Success is rare Compare observed and expected frequency Probabilities of all outcomes sum to 1 Outcomes are mutually exclusive
For a given treatment, the ratio of the probability of success to the probability of failure is called the ___________.
Odds
Over the course of 14 days, I recorded the number of times my dog barked when the sound of a doorbell came from the tv. She barked 12 times out of the 19 times a doorbell was heard on the tv. What are the odds of you dog barking at a tv doorbell?
Odds = [A]/[B] Odds = [12/19]/[7/19]
One sided test
Only one meaningful outcome
Results of a fixed-effects meta-analysis allow inference to _______.
Only those studies included in the meta-analysis.
Which of the following best describes the shape of a distribution with positive skewness?
Peak is at the lower end and there is a long tail trailing at the upper end
Which of the following are benefits of rigorous data management?
Peers can evaluate legitimacy of data Scientific community and general public have more confidence in results published from well-managed data Data are more likely to be free of errors Data can be used by someone else in the future
If a researcher conducts an experiment with 1 fixed factor consisting of 5 treatments and identifies particular interest in comparing means between treatments 2 and 4 before conducting the experiment, what type of comparison of those two treatments would they do if their overall ANOVA is significant?
Planned
Kurtosis
Pointings of a peak Positive: very pointy, high Y values Negative: flat, low Y values Normal: 3, excess kurtosis = 0
For each consideration listed below, indicate the probability distribution to which it applies. 1: distribution assumes the probability of success is low.
Poisson distribution only
In a phase 1 study testing the efficacy of a new drug therapy for heart disease, researchers randomly select 1000 participants with the targeted type of heart disease. Of these, 500 currently admitted to the hospital are assigned to the treatment group receiving a pill with the drug. The other 500, who are stable and at home, are given a placebo pill that does not contain the drug. What is the source of bias?
Presence of confounding variables or lack of blinding
What is this type of plot used to evaluate when used in the context of meta-analysis?
Publication bias
You conduct an experiment of randomly selected patients having a range of low density lipoprotein (LDL) cholesterol concentrations and monitored their blood pressure to examine if LDL cholesterol causes higher blood pressure. What type of analysis would be most appropriate?
Regression
Which of the following analyses could be used to assess linear relationships between two continuous numeric variables?
Regression Correlation
An experiment examining corn resistance to insects using a new herbicide is conducted with 10 plots randomly assigned to be treated with the herbicide as a spray and 10 plots randomly assigned to be sprayed with water vapor only. During the experiment, a combine harvests the corn prior to the researchers completing data collection so that only 8 control plots remain while all 10 treatment plots remain throughout the data collection process. Which of the following statements is most true?
Researchers will be able to analyze the data, but statistical tests will have less power due to the lack of equal sample sizes in the treatment and control groups in the final dataset.
An experiment examining corn resistance to insects using a new herbicide is conducted with 10 plots randomly assigned to be treated with the herbicide as a spray and 10 plots randomly assigned to be sprayed with water only. During the experiment, a combine harvests the corn prior to the researchers completing data collection so that only 8 control plots remain while all 10 treatment plots remain. Throughout the data collection process. Which of the following statements is most true?
Researchers. Will be able to analyze the data, but statistical tests will have less power due to the lack of equal sample sizes in the treatment and control groups in the final dataset.
Metadata provides complete and accurate documentation of the life cycle for a dataset so that a peer researcher can use it appropriately in a new data-mining project
Reusable
Which of the following are assumptions of the non-parametric equivalent of a one-way ANOVA?
Sample units were randomly drawn form the statistical population Samples have equal variances Sample units are independent of one another
Which of the following are assumptions of the non-parametric equivalent of a one-way ANOVA?
Samples have equal variances Sample units were randomly drawn from the statistical population Sample units are independent of one another
Which of the following are benefits of rigorous data management?
Scientific community and general public have more confidence in results published from well-managed data Data are more likely to be free of errors Peers can evaluate legitimacy of data Data can be used by someone else in the future
Binomial Distribution Assumptions
Set number of independent trials Exactly 2 mutually exclusive possible outcomes per trial (heads or tails) Probability of success plus failure =1
Which of the following tests can be used to evaluate whether samples meet the assumption of normality?
Shapiro-Wilk Test
Which of the following tests can be used to evaluate whether y-residuals are normally distributed?
Shapiro-Wilk test
Researchers wanted to know if beak size in Gouldian finches differed between males and females. They measured beak width and length on 25 female and 25 male Gouldian finches. On average, female beaks were 10.2 mm wide and male beaks were 9.7 mm wide. They conducted a Shapiro-Will test and found: W-statistic = 2.298, p-value = 0.127. They also conducted a Levene's test and found: F-statistic = 0.374, p-value: 0.0443. Based on this information, what is the next step researchers should take in the data analysis process and why?
Shapiro-Will (0.127) > 0.05, so normal. Leven's p-value (0.0443) < 0.05, so variance is not equal. So, should proceed with Welch's t-test.
If a sample is strong positive skew, which non-parametric test would be most applicable?
Sign test
Which of the following are directly or indirectly required in order to calculate the upper and lower confidence limits for a normal distribution?
Significance level, sample size, t-critical, sample mean
The null hypothesis of regression analysis is _________.
Slope = 0
Which of the following are measures of variation or spread, but not uncertainty, that can be calculated from sample data?
Standard Deviation and Interquartile Range
In reporting the results of a two-sample experiment, which of the following should you include in the report to make it possible or someone else to use your study as part of a meta-analysis?
Standard Errors of each group Test statistic Degrees of freedom Means of each group Sample size
Coefficient of Variation
Standard deviation divided by the mean
Odds = 1
Success and failure are equally likely
Odds < 1
Success is less likely
Odds > 1
Success is more likely
Sum of squares
Sum of the squared deviations
The test statistic for a correlation analysis is:
T-statistic
For the characteristic provided, identify if it applies to t-tests only, analysis of variance (ANOVA) tests only, both t-tests and analysis of variance tests, or neither t-tests nor ANOVA. Can test the null hypothesis that 1 sample mean differs from a specific hypothesized value.
T-test only
A journal article reports that they used alpha = 0.05 in the statistical analysis I portion of the methods section. In presenting results for a correlation matrix in which 10 continuous variables were screened, they refer to an Bonferroni alpha' (alpha prime). What can you infer from this?
The adjusted alpha should be less than 0.05. They adjusted alpha for individual correlations so that across all correlation tests they maintained alpha. =0.05.
Interquartile range
The difference between the upper and lower quartiles.
You read the following conclusion from a correlation analysis: COVID-19 mortality was correlated with air borne particulate matter concentrations (PM2.5) (r=0.55, t=2.75, DF=51, p=0.0457). Based on the information provided, what can you conclude about the relationship between the two variables?
The null hypothesis was rejected As PM2.5 increases, so does mortality from COVID-19.
Statistical populations
The objects we want to make inferences on
A study was done to assess the odds of seeing a wolf in Yellowstone National Park in May, June, and July. Which of the following represents the most plausible interpretation for an odds ratio of 1.7 and 95% confidence interval of 0.9 to 2.5 for the probability of seeing a wolf in May relative to June.
The odds of seeing a wolf in June is the same as the odds of seeing a wolf in May.
A study was done to assess the odds of seeing a wolf in Yellowstone National Park in May, June, and July. Which of the following represents the most plausible interpretation for an odds ratio of 3.1 and 95% confidence interval of 1.9 to 3.5 for the probability of seeing a wolf in May relative to June.
The odds of seeing a wolf in May is higher than the odds of seeing a wolf in June.
Sample
The unit to which a treatment is applied
After analyzing the data, the statistical test results in a p-value of 0.062.
There is not a statistically significant difference between groups Fail to reject the null hypothesis
Which of the following terms is calculated as the number of sample units in the study minus one?
Total degrees of freedom
Which of the following terms relates to the difference between an individual observation and the grand mean?
Total deviation
A one-sample t-test is conducted to a evaluate whether a population mean differs from a hypothesized mean
True
A one-sample t-test is conducted to evaluate whether a population mean differs from a hypothesized mean
True
A p-value indicates the probability of getting a test statistic as extreme or more extreme assuming the null hypothesis is true
True
Both the chi-square goodness-of-fit test and the chi-square contingency analysis assume that non of the mutually exclusive categories have an expected frequency less than 1.
True
Chi-square tests compare observed frequencies to expected frequencies. Expected frequencies can be calculated from any theoretical probability distribution
True
Chi-square tests compare observed frequencies to expected frequencies. Expected frequencies can be calculated from any theoretical probability distribution.
True
Correlation analysis can be used to examine whether there is an association between two continuous numeric variables.
True
If data are not normally distributed, a data transformation may be able to improve normality.
True
If data are not normally distributed, a data transportation may be able to improve normality
True
In addition to the standard assumptions for general linear models, ANVOCA analysis also assumes that the covariate will not be affected by the treatment?
True
The Mann-Whitney U-test assigns ranks to values blind to treatment designation so that there is only one value assigned to each rank across both samples
True
The binomial test would be appropriate for testing the following null hypothesis: The probability of winning the lottery is less than 0.1%.
True
The null hypothesis of the chi-square goodness-of-fit test is that the frequency distribution observed in a sample does not differ from the frequency expected under a given theoretical probability distribution.
True
Two sided test
Two possible outcomes, default to this test
Researchers wanted to know if beak size in Gouldian finches differed between males and females. They measured beak width and length on 25 female and 25 male Gouldian finches. To test whether beak width differed between females and males, which test would be most appropriate?
Two sample t-test
In an experiment testing whether two drugs interact with one another to influence patient outcomes, which one of the following experimental designs would be most appropriate tot use for statistical analysis?
Two-way factorial analysis of variance
Ruff et. Al. Examined whether Neanderthal and early modern humans exhibited significant differences in cranial capacity. However, they wanted to account for the overall larger body size of Neanderthals compared to early humans. Based on these results, which of the following conclusions is most appropriate?
Unable to draw a final conclusion because an equal slopes model without the species*mass interaction would be most appropriate.
When screening correlations among a set of 10 continuous variables, which of the following adjustments could be used to maintain the overall Type I error rate?
Use a Bonferroni adjustment to alpha Use a Dunn-Sidak adjustment to alpha
Which of the following best represents the null hypothesis of the Levene's test?
Variances of two samples are equal
Which of the following are disadvantages of a study design that produces paired data
Violates the assumption of independence between sample means
Accurate, precise
What it will be if dots are close together in bulls eye
Bias
When a sample is precise but not accurate
Which of the following general linear models is most likely to represent an equal slopes analysis of covariance (ANCOVA)?
Y = mean + factor A + covariate + error
Sampling distribution
Y-axis is labeled as the frequency of Y-bar,or the frequency of sample averages will converge on the statistical. Population's mean
The value of Y when X=0 is called the __________.
Y-intercept
The value of Y when X=0 is called the:
Y-intercept
Statistical population
a group of similar things that a scientist is interested in learning about
Precision
a measure of how close a series of measurements are to one another
Blinding
a technique where the subjects do not know whether they are receiving a treatment or a placebo
Ordinal Data
a type of data that refers solely to a ranking of some kind
Frequency distribution
an arrangement of data that indicates how often a particular score or observation occurs
Two-tailed hypothesis
both directions of an effect or relationship are considered in the alternative hypothesis of the test There is a difference between the responses of treatment and control groups
Abductive Reasoning
concluding something is true by testing hypotheses with evidence Spiders eat mosquitos. Mosquitos are in my back yard. Conclusion: spiders are in my back yard.
Numerical Data
discrete and continuous
Randomized complete block design
each block has an equal and complete number of treatments, which are randomly distributed within each block
All Z tables only provide probabilities for the area under the curve from a critical value to the upper tail (Pr[Z > critical value]) of a standard normal distribution.
false
Type II error
false negative
Type I error
false positive, significance level
Interquartile range
middle 50% of data 25% above 25% below median
Categorical Data
nominal and ordinal
One-tailed hypothesis
only one direction of an effect or relationship is predicted in the alternative hypothesis of the test Greater than or equal to zero OR less than zero
Accuracy
the closeness of a measurement to the true value of what is being measured
Deductive Reasoning
the process of applying a general statement to specific facts or situations Spiders have 8 legs. Tarantulas are spiders. Conclusion: tarantulas have 8 legs.
Standard deviation
the square root of the variance
Variance
the sum of squared deviations from the mean, divided by the count minus one standard deviation squared
