stats for ANE

what % of the data fall within 1SD? 2?3?

+-1, 68% +-2, 95% +-3, 99.7%

Which of below is NOT a descriptive statistic? - Stem plot of data - Skewness of a data distribution - Confidence interval derived from data - Spread of a data distribution

- Confidence interval derived from data

Which statement is NOT correct for p-value? - p-value lower than the cutoff value will lead to rejection of null hypothesis - p-value can be -0.09 - p-value is a probability

- p-value can be -0.09 It cannot be negative

calculate z score

1. subtract # by the mean 2. divide by s.d. z= (y-u)/σ

what is the central limit theorem? (3)

1. the random sampling distribution of means will always tend to be normal, irrespective of the shape of the population distribution from which the samples were drawn. 2. The random sampling distribution of means will become closer to normal as the size of the sample increases 3. the mean of the random sampling distribution of means is equal to the mean of the original population

In 2016, the Pew Research Center conducted a survey of 1,003 Americans to assess their opinion of climate change. The collected data show 44% of Americans believe climate change is a very serious problem. The estimated 95% CI of the proportion is [39%, 49%]. What is the point estimate of proportion of Americans who believe climate change is a very serious problem?

44%

You have collected data about the average response time of participants in your study. You are delighted to find that the variable is normally distributed. The mean response time is 30 seconds and the standard deviation is 4 seconds what percent of your subjects would you expect to respond in 26-34 seconds? A. 50% B. 68% C. 95% D. Unable to determine

68%

what is the confidence interval? n=500 SD = 1.5 x=7.5 CL = 95% = z score = 1.96

7.5-+(1.96(1.5/√500)) 7.5 -+(0.1372) 7.36 <= u<= 7.64

Quota sampling

A form of nonprobability sampling in which you select the proportions of the sample for different subgroups, much the same as in stratified sampling but without random selection. (50% of subjects are male and 50% are female)

Convenience sampling

A form of nonprobability sampling that consists of collecting data from the group that is available. (first 6 patients are sampled that meet criteria)

Your unit completes the internal controls module at your workplace. Two of the nurses on your unit were internal control specialists in their last place of employment and scored a perfect score on the exam while the rest of the nurses had an average score of 78 with a range of 72-84 on the test. When you look at the whole grade distribution you are not surprised to see: A. A positive skew B. A negative skew C. A normal distribution D. The mean, median and mode are equal

A positive skew

Sampling bias

A systematic error made in the sample selection that results in a nonrandom sample.

You read a large study that indicates taking nitrofurantoin is associated with hepatic impairment. However, after examining the sampling methodology and the statistics you believe this conclusion is an error. If the conclusion is an error you know this means:

A type I error was made (false positive)

Sampling distribution

All the possible values of a statistic from all the possible samples of a given population.

what are the requirements of a causal inference?

Causal inferences can be drawn from randomized experiments Causal inferences cannot be drawn from observational studies due to Confounding variable(s)

convergent validity

Comparing the results from a new instrument with an existing instrument that measures the same thing If the correlation between the two instruments is 0.4 or higher it strengthens the validity of both instruments

Confidence Level

Confidence Level is the likelihood, expressed as a percentage, that a specified interval will contain the population parameter Common confidence levels: 90%, 95%, 99%

Descriptive statistics

Descriptive statistics are procedures that help us summarize and describe data collected from either a sample or a population. mean, SD, z score charts and graphs, shape

Sampling error

Differences between the sample and the population that occur due to randomization or chance.

what is efficiency? EFF

EFF = A+D(#accurate tests)/(total sample) the portion of the sample that has true positives and negatives divided by the total.

Patients on your unit complete a functional mobility test. The test determines how many seconds it takes to walk 20 feet unassisted. Your patient's speed is reported as a Z score of 2.3. Her roommate's speed is reported as a Z score of 1.8. You would expect your patient to walk 20 feet unassisted: A. Unable to determine B. Faster than her roommate C. Slower than her roommate D. 2.3 times faster than the slowest patient on the unit

Faster than her roommate

what is another name for the normal distribution?

Gaussian distribution the dudes name

Your patient returns from physical therapy with a summary report that states her post-surgical range of motion has a corresponding Z score of -0.43. You know this means your patient:

Has worse range of motion when compared to the average post-surgical patient.

High sensitivity is good at what?

High sensitivity Good at identifying the diseased patient Low chance of false negatives Very important if a disease is fatal, contagious or early treatment helps

High specificity

High specificity Good at identifying the patients without a disease Low chance of false positives

2,000 adults between 20-30 years old, were randomly collected from the target population suffering from constant tiredness. Then the participants were randomly divided into 2 groups, with 1000 participants in each. One group of participants (the intervention group) were given a new drug to treat the fatigue. The control group of participants was given a placebo pill. Neither the participants nor the researchers involved in the experiment knew whether they were taking the new drug or the placebo. The participants took the pills for 3 weeks, 2 per day. A scale is used to measure participants' levels of tiredness. A 95% confidence interval, , of mean difference of tiredness scores for the two groups was derived from the data. Which of the following strategies can help if the researchers want to improve the precision of CI?

Increase the number of patients in the sample from 2,000 to 3,000.

What are the requirements of a inference that can be generalized?

Inference can be generalized to population if units are a Random Sample from that population. Inference cannot be generalized to population if units are self-selected, e.g. in this creativity example, inference can only be drawn to the subjects in the sample that was taken.

Inferential statistics

Inferential statistics are the logic and procedures concerned with making predictions or inferences about a population from observations and analyses of a sample - hypothesis testing and Estimation and confidence interval

if the data is negatively skewed is the mean or median higher? left or right?

Left skewed, mean<median

Margin of error (ME)

Margin of error (ME) is the radius of a confidence interval

Nonprobability sampling

Methods in which subjects do not have the same chance of being selected for participation (not randomized).

2,000 adults between 20-30 years old, were randomly collected from the target population suffering from constant tiredness. Then the participants were randomly divided into 2 groups, with 1000 participants in each. One group of participants (the intervention group) were given a new drug to treat the fatigue. The control group of participants was given a placebo pill. Neither the participants nor the researchers involved in the experiment knew whether they were taking the new drug or the placebo. The participants took the pills for 3 weeks, 2 per day. A scale is used to measure participants' levels of tiredness before and after the trial. The study result shows that there is a significant decrease in fatigue level for the new drug group. What can we NOT say about the study?

NOT: A causal inference can be drawn that the new drug has an effect on the tiredness level.

You are reading a study that examines the association between taking levothyroxine and the therapeutic insulin dose in diabetics. The researchers report a p=0.01. You know this means that:

NOT: There is a clinically significant association.

2,000 adults between 20-30 years old, were randomly collected from the target population suffering from constant tiredness. Then the participants were randomly divided into 2 groups, with 1000 participants in each. One group of participants (the intervention group) were given a new drug to treat the fatigue. The control group of participants was given a placebo pill. Neither the participants nor the researchers involved in the experiment knew whether they were taking the new drug or the placebo. The participants took the pills for 3 weeks, 2 per day. A scale is used to measure participants' levels of tiredness. A 95% confidence interval, , of mean difference of tiredness scores for the two groups was derived from the data. Which statement is true about this confidence interval?

NOT: We are 95% confident that the difference of the sample mean tiredness scores between the two groups is between 1 and 5.

What is NPV?

NPV = D(true neg)/C+D(total # tested neg) probability of not having the disease if the result is negative

Researchers study the impact of internalized bias on patient care. Nurses complete a survey which determines a score for their internalized bias from 0-100. Without knowing the internalized bias score, patients are asked to rank the subsequent care they receive from the nurse as poor, fair, good or exemplary. In this study what type of variable is your independent variable?

Numeric

What is PPV?

PPV = A(true pos)/A+B(total # tested pos) Positive Predictive Value (PPV): the probability of a subject having the disease given a positive test result

what happens to PPV when prevelance goes up?

PPV goes up too

What is prevelance?

Prevalence = A+C(all w/ dx)/A+B+C+D(total sample) The proportion of persons who have a condition at or during a particular time Is a measure of disease burden Prevalence (estimated)= (A+C)/N = (All with disease)/(Entire sample)

Simple random sampling

Probability sampling in which every subject in a population has the same chance of being selected.

Systematic sampling

Probability sampling involving the selection of subjects according to a standardized rule.

Stratified sampling

Probability sampling that divides the population into subsamples according to a characteristic of interest and then randomly selects the sample from these subgroups.

Cluster sampling

Probability sampling using a group or unit rather than an individual.

calculate IQR

Q3-Q1

standard error and equation?

SEM measures how far the sample mean of the data is likely to be from the true population mean. The SEM is always smaller than the SD. SEM= SD/√n

What is sensitivity?

SENSITIVITY = A(true pos)/A+C(all who have dx) Probability of being tested positive if having that disease (the ability of a test)

What is specificity?

SPECIFICITY = D(true negs)/B+D(all that do not have Dx) The probability that the patient will test negative for the disease if the patient does not have the disease (the ability of a test)

Probability sampling

Techniques in which the probability of selecting each subject is known (randomized).

Which statement is true about the normal curve? The total area under the curve is 50% The area under the curve represents the probability A normal curve is usually constructed using our collected data None of above is correct

The area under the curve represents the probability the area under the curve is 100%

Inclusion criteria

The list of characteristics a subject must have to be eligible to participate in a study.

Exclusion criteria

The list of characteristics that would eliminate a subject from being eligible to participate in a study.

The hypothesis testing results of a study reports that administering vancomycin incorrectly is associated with red man syndrome. This means:

The p value was less than alpha.

Sampling method

The processes employed to select the subjects for a sample from the population being studied.

A study reports that there is no association between taking metronidazole and sun sensitivity. You know this means:

The result is not statistically significant.

You have collected data about the post void residuals (PVR) of participants in your study. The mean PVR is 120 mls. Because you have some outliers noted in your study you also report a median PVR which is 138 mls. Based on this data you know: A. There is at least one outlier that is on the left side of the distribution B. The researcher is biased C. The outliers lie in equal amounts on both sides of the distribution D. There is a bimodal distribution

There is at least one outlier that is on the left side of the distribution

how do you describe the skew of the data? concerning mean median, direction, pos and neg.

When the mean is higher than the median, the distribution is positively/right skewed. (right tail) When the mean is lower than the median, the distribution is negatively/left skewed (left tail)

You know the following is true when conducting a hypothesis test.

You should determine your significance level or alpha before conducting your statistical test.

divergent validity

a measure that is uncorrelated with different measures (correlation of 0) A correlation of at least -0.4 (strong negative correlation) strengthens the divergent validity of both measures

Confidence Interval (CI)

a range of values defined by the confidence level within which the population parameter is estimated to fall

what is a confounding variable?

a variable that influences both independent variable (e.g. gender) and dependent variable (e.g. salary), thus distorting the association.

You have collected data about the average age of participants in your study. You are delighted to find that the variable is normally distributed. If the mean age is 12 years and the standard deviation is 4 years. What is the median age for your subjects? A. 8-16 years B. about 12 years C. Unable to determine D. Greater than 12 years

about 12 years

what is the difference between cluster and stratified sampling method?

cluster sampling: break the population into groups and randomly select SOME of the groups. your sample is everyone in the groups sampled (everyone in five 7/11's) Stratified: break population into groups then randomly select INDIVIDUALS from EACH group. (2 people from each 7/11)

what is the difference between cluster and stratified sampling?

cluster: Doesn't require a list of individuals, SRS clusters then select according to a scheme, cheap, need a lot of data to be precise. stratified: Requires more information, Ensures representation of groups across the population,

What is reliability?

consistency of measurement precision not accuracy

what are the major categories of non-probability sampling? (5)

convenience quota network judgmental/purposive volunteer

What is the key step of statistical process before we can design study and collect data to investigate a research question?

create a hypothesis

what is the difference between generalization and causality?

generalization is pertaining to how the sample is gotten (random vs selected/volunteered), while the causality is how your study will be then tested (random vs observational) so if hospitals are randomly selected to study staffing ratios this can be generalized but not causal because the hospitals are randomized, but the staffing ratios are preassigned and cannot be randomized.

what are the levels of variables?

highest to lowest: ratio - has an absolute 0 (height weight, temp in Kelvin) interval - distance is meaningful (and measured at a specific interval) ordinal - attributes can be ordered nominal - attributes are only named: weakest

Point Estimate

is a sample statistic used to estimate the exact value of a population parameter

what is the JHNEBP hierarchy?

level 1: randomized controlled trial (RCT), explanatory mixed methods with only level I quantitative study, or systematic review of RCTs 2: A quasi-experimental study, explanatory mixed methods with only level II quantitative study, or systematic review of a combination of RCTs and quasi- experimental studies, or quasi-experimental studies only, with or without meta- analysis 3: A quantitative non experimental study, combination of studies.

what measures of central tendency (average) can be used with ordinal data?

median and mode

T/F: 225 hospitals were randomly selected from a target location to compare the high and low patient-to-nurse staffing ratios in all-cause readmission among children for common medical and surgical conditions. The study found a significant difference in the readmission between high and low staffing ratios. A causal inference can be drawn from this study.

no, because this is an observational study (correlation not causation)

Researchers study the impact of internalized bias on patient care. Nurses complete a survey which determines a score for their internalized bias from 0-100. Without knowing the internalized bias score, patients are asked to rank the subsequent care they receive from the nurse as poor, fair, good or exemplary. In this study what is your dependent variable?

ordinal

The total set of cases, objects, groups, or events in which the researcher is interested is called the

population

what are the two types of stratified random sampling?

proportional - where the size of the group will effect the amount of subjects selected disproportionate - the size of the sample selected from each subgroup is disproportional to the size of that subgroup in the population

A survey asks your patient to identify his primary language. The choices are:1.) Spanish2.) English3.) Arabic4.) OtherYou know this is an example of what type of variable?

qualitative

how do you calculate a confidence interval?

range from [mean-Z(SEM)] to [mean+Z(SEM)] 95% CI z score is z=1.96 n=7 x= 29.86 u= SD=7.08 29.86 - 1.96(7.08/√7)<=u<= 29.96 + 1.96(7.08/√7)

A study examines the relationship between educational preparation and scores on a cultural competency exam. Subjects included are nurses with an associate's degree, nurses with a baccalaureate degree, nurses with a master's degree, and nurses with a doctoral degree. In this example, cultural competency is measured at what level?

ratio

what are the categories of probability sampling?

simple random sample stratified (divide into groups then SRS) systematic random (K=N/n.... every Kth chosen) cluster there is a know probability of each

The researchers wants to investigate the association between age and hearing impairment. They collect a sample of patients. The ages are recorded and the variance is 18. What is the standard deviation of the age?

square root of variance is SD= 4.2

what are the three types of consistency when measuring a specific variable?

stability: test- retest reliability homogeneity: internal consistency equivalence: inter-rater reliability and parallel-forms reliability

what are the two types of error? not 1 and 2.

systematic and random random cannot be eliminated just decreased systemic is from faulty equipment or experiment design.

What is kurtosis?

the "peakedness" of the distribution positive is tall negative is flat

What is internal validity?

the degree to which data in a study reflect a true cause-effect relationship. same as causal inference.

what is error?

the difference between the experimental value and the accepted value e = O-T

What is Chronbach's alpha?

the reliability metric that should reflect the degree of internal consistency of test items measuring the same skill. the >0.4 thing

what is type 1 and 2 error?

type 1 false positive (male w/ positive pregnancy test) type 2 false negative (pregnant female with a negative test)

T/F: 225 hospitals were randomly selected from a target location to compare the high and low patient-to-nurse staffing ratios in all-cause readmission among children for common medical and surgical conditions. The study found a significant difference in the readmission between high and low staffing ratios. The inference can be generalized.

yes the inference can be generalized

how many SD away from the pop mean is ybar? SD = 30 Ybar = 180 u = 150 n = 100

z = (y-u)/σ = (y-u)/(σ/√n) = (180-150)/(30/√100) = 10 the probability that 180 is the observed mean from sampling 100 people from the population is basically zero.