click below
click below
Normal Size Small Size show me how
Biostatistics FA26
| Question | Answer |
|---|---|
| What is mean? | Add all numbers and divide by the number of observations |
| What is median? | The middle value after sorting data. For an even count, average the two middle values. |
| What is variance? | How spread out values are around the mean; it uses squared deviations. |
| What is standard deviation? | The square root of variance, measuring spread in the original units. |
| What is Interquartile Range? | Q3 − Q1; the spread of the middle 50% of observations. |
| How do you calculate mean? | |
| How do you calculate median? | |
| How do you calculate standard deviation? | |
| How do you calculate variance? | |
| What is an outlier? | |
| How does an outlier influence measures of central tendency? | |
| What is the shape of a normal distribution? | Bell curve |
| The tail of a curve points left, this is what type of skew? | Right |
| The tail of a curve points right, this is what type of skew? | Left |
| What is the mode? | The most frequently occurring value in a dataset. |
| How do you calculate sample variance? | Sum squared deviations from the mean, then divide by n − 1. |
| What is the difference between population and sample variance? | Population variance divides by N; sample variance divides by n − 1. |
| What is an outlier, and how can it affect the mean? | An unusually distant observation; it can pull the mean toward itself. |
| Which is more resistant to outliers: mean or median? | Median; it depends on the middle position rather than extreme values. |
| the mean is often greater than the median. Left or Right Skew? | Right |
| the mean is often less than the median. Left or Right Skew? | Left |
| In a normal distribution, what falls within 1 SD of the mean? | Approximately 68% of values. |
| In a normal distribution, what falls within 2 SDs of the mean? | Approximately 95% of values. |
| In a normal distribution, what falls within 3 SDs of the mean? | Approximately 99.7% of values. |
| What is a quantitative variable? | A numerical measurement or count, such as age, weight, or number of visits. |
| What is a categorical variable? | A variable representing groups, such as blood type or treatment group. |
| What is a time-to-event variable? | The time until an event occurs, such as time until relapse or death. |
| What is a cross-sectional study? | Measures exposure and outcome at one point or period in time. |
| What is a cohort study? | Follows groups over time to compare outcome occurrence by exposure. |
| What is a case-control study? | Starts with outcome status and looks back to compare prior exposures. |
| What is a randomized controlled trial? | Researchers randomly assign interventions and compare outcomes between groups. |
| What is incidence? | The occurrence of new cases in a population over a specified period. |
| What is cumulative risk? | The proportion initially at risk who develop the outcome during a specified period. |
| What is prevalence? | The proportion of a population that has a condition at a given time or period. |
| What is an incidence rate? | New cases divided by total person-time at risk. |
| What is an absolute risk difference? | Risk in one group minus risk in another group. |
| What does number needed to treat (NNT) mean? | People who must receive treatment to prevent one additional adverse outcome; NNT = 1/absolute risk reduction. |
| What is the range of a dataset? | Maximum value minus minimum value. |
| What does a high SD indicate? | Values are more spread out around the mean. |
| What does a low SD indicate? | Values tend to cluster closer to the mean. |
| What is the difference between discrete and continuous data? | Discrete data are countable; continuous data can take any value within a range. |
| Is blood type quantitative or categorical? | Categorical: it identifies groups rather than numerical amounts. |
| Is number of clinic visits discrete or continuous? | Discrete quantitative: it counts whole visits. |
| What is a confounder? | A third variable that can distort the observed exposure–outcome relationship. |
| Observational vs. experimental study: what's the difference? | Observational studies don't assign exposure; experimental studies assign an intervention. |
| What is the main benefit of random assignment? | It helps balance confounders across groups, strengthening causal inference. |
| Incidence vs. prevalence: what's the difference? | Incidence counts new cases; prevalence counts all existing cases. |
| What is person-time? | Total time participants contribute while at risk and under observation. |
| Does prevalence include old and new cases? | Yes. It includes all existing cases in the population being measured. |
| What does NNH stand for? | Number needed to harm: 1 divided by the absolute increase in risk. |
| When is NNT calculated using absolute risk reduction? | When treatment reduces risk: NNT = 1/ARR, rounded up to a whole number. |
| Why must cumulative risk specify a time period? | The probability of developing an outcome depends on the follow-up duration. |
| How do mean, median, and mode relate in a normal distribution? | They are equal or approximately equal at the center. |
| What is a major limitation of cross-sectional studies? | They often cannot establish whether exposure occurred before the outcome. |
| What does an ARR of 0.10 mean? | Treatment reduces absolute risk by 10 percentage points. |
| What is the number needed to treat (NNT) formula? | NNT = 1 ÷ ARR, using ARR as a proportion. |
| ARR is | Absolute Risk Ratio |
| What is the number needed to harm (NNH) formula? | NNH = 1 ÷ absolute risk increase, using a proportion. |
| Treatment risk = 15%, control risk = 10%. What is NNH? | ARI = 0.05; NNH = 1 ÷ 0.05 = 20. |
| What is an incidence-rate difference? | Incidence rate in one group minus the rate in the comparison group. |
| Treatment rate = 4; control rate = 9 cases per 1,000 person-years. Difference? | 4 − 9 = −5 cases per 1,000 person-years; the treatment rate is lower. |
| What is a risk ratio (RR)? | Risk in the exposed or treatment group divided by risk in the comparison group. |
| Treatment risk = 20%, control risk = 10%. Calculate RR. | RR = 0.20 ÷ 0.10 = 2. |
| How do you interpret RR = 1? | The groups have equal risks; there is no relative difference in risk. |
| How do you interpret RR = 2? | The first group's risk is twice the comparison group's risk. |
| How do you interpret RR = 0.70? | The first group's risk is 30% lower than the comparison group's risk. |
| How do you convert RR into percent increase? | (RR − 1) × 100%, when RR > 1. |
| How do you convert RR into percent decrease? | (1 − RR) × 100%, when RR < 1. |
| RR = 1.40. What is the relative increase in risk? | (1.40 − 1) × 100% = 40% higher risk. |
| What is a rate ratio? | Incidence rate in one group divided by incidence rate in another group. |
| Treatment rate = 12; control rate = 4 per 1,000 person-years. Rate ratio? | 12 ÷ 4 = 3; the treatment group's incidence rate is three times as high. |
| What is a prevalence ratio? | Prevalence in one group divided by prevalence in another group. |
| What is an odds ratio (OR)? | Odds of the outcome in one group divided by odds in the comparison group. |
| How do you calculate odds from a probability p? | Odds = p ÷ (1 − p). |
| If risk is 20%, what are the odds? | 0.20 ÷ 0.80 = 0.25, or 1:4. |
| Cases: 40 exposed, 60 unexposed; controls: 20 exposed, 80 unexposed. Odds Ratio | (40 × 80) ÷ (60 × 20) = 2.67. |
| What is a z-score? | The number of standard deviations a value is above or below the mean. |
| What is the formula for a z-score? | z = (individual value − mean) ÷ standard deviation. |
| A value is 80, the mean is 70, and SD is 5. Calculate z. | z = (80 − 70) ÷ 5 = 2. The value is 2 SDs above the mean. |
| What does a z-score of 0 mean? | The value equals the mean. |
| What does a positive z-score mean? | The value is above the mean. |
| What does a negative z-score mean? | The value is below the mean. |
| What is covariance? | A measure of how two variables vary together. |
| What is correlation? | A standardized measure of the strength and direction of a relationship between two variables. |
| What is the range of the correlation coefficient r? | −1 to +1. |
| What does r = +1 mean? | A perfect positive linear relationship. |
| What does r = −1 mean? | A perfect negative linear relationship. |
| What does a positive correlation mean? | Higher values of one variable tend to occur with higher values of the other. |
| What does a negative correlation mean? | Higher values of one variable tend to occur with lower values of the other. |
| What does r = 0.80 suggest? | A strong positive linear relationship, using common descriptive guidelines. |
| What does r = −0.20 suggest? | A weak negative linear relationship, using common descriptive guidelines. |
| What is an outlier's possible effect on correlation? | An outlier can substantially increase, decrease, or otherwise distort the correlation. |
| A value has z = 1.5. How should you interpret it? | It is 1.5 standard deviations above the mean. |
| What is probability? | A number from 0 to 1 describing how likely an event is. 0 = impossible; 1 = certain. |
| What is the formula for theoretical probability when outcomes are equally likely? | P(event) = number of favorable outcomes ÷ total number of possible outcomes. |
| What is the probability of flipping heads on a fair coin? | 1/2, 50% |
| What is the probability of rolling a 4 on a fair six-sided die? | 1/6 ≈ 0.167, or 16.7%. |
| What is the probability of rolling an even number on a fair six-sided die? | 3/6 = 1/2 = 50%, because 2, 4, and 6 are even. |
| A standard deck has 52 cards. What is the probability of drawing an ace? | 4/52 = 1/13 ≈ 7.7%. |
| What is empirical probability? | Probability estimated from observed data: number of times an event occurred ÷ total number of observations. |
| A study observes 30 positive results among 100 people. What is the empirical probability of a positive result? | 30/100 = 0.30, or 30%. |
| What is the probability of getting heads twice in two independent fair coin flips? | 1/2 × 1/2 = 1/4 = 25%. |
| What does 'at least one' mean in probability questions? | One or more occurrences. It is often easiest to calculate 1 − P(none). |
| What is the probability of at least one head in two fair coin flips? | 1 − P(no heads) = 1 − 1/4 = 3/4 = 75%. |
| How do you calculate standard deviation from variance? | Take the square root of the variance: SD = √Var(X). |
| If a random variable has variance 9, what is its standard deviation? | SD = √9 = 3. |
| A test has a 10% false-positive probability for each independent test. What is the probability of at least one false positive across two tests? | 1 − P(no false positives) = 1 − (0.90 × 0.90) = 0.19, or 19%. |
| What is a binomial distribution? | A probability distribution that counts the number of successes in a fixed number of trials, when each trial has two possible outcomes. |