stats class 4

What is quota sampling?

A sampling method in which the researcher sets subgroup targets (quotas) that mirror the population and fills each quota with whoever is convenient to reach.

How are quotas determined in quota sampling?

The researcher sets targets for subgroups to mirror the population.

How are participants selected in quota sampling?

Participants are conveniently recruited until each subgroup quota is filled.

In quota sampling, what is the key distinction from stratified sampling?

Selection within each quota is not random.

What is an example of quota sampling from the lecture?

A unit needs 60% female nurses and 40% male nurses surveyed, so recruitment continues until both quotas are filled.

When is quota sampling useful?

When speed and low cost are important and rough representativeness matters more than statistical rigor.

What is a major limitation of quota sampling?

It is vulnerable to selection bias because convenient participants within a quota may not represent that subgroup well.

What does alpha (α) measure?

The probability of rejecting H₀ when H₀ is actually true.

What is a Type I error?

Rejecting H₀ when it is actually true; a false positive.

What is the usual alpha threshold in the lecture?

0.05.

When is alpha set?

Before the study, before data are collected.

What does beta (β) measure?

The probability of failing to reject H₀ when H₀ is actually false.

What is a Type II error?

Failing to reject H₀ when it is actually false; a false negative.

What is statistical power?

The probability of correctly detecting a real effect if one truly exists.

What is the formula for power?

Power = 1 − β.

What is the relationship between power and beta?

Power equals one minus beta; lowering beta increases power.

How are alpha and beta related when sample size is fixed?

They involve a trade-off; lowering alpha pushes beta up, meaning more real effects may be missed.

How can both alpha and beta be improved at once, according to the lecture?

Increase the sample size.

What does alpha represent in the alpha-versus-p-value analogy?

The passing grade decided on in advance.

What does the p-value represent in the alpha-versus-p-value analogy?

The score produced by the exam.

What is alpha (α) in relation to the p-value?

A threshold set before data collection, most often 0.05.

What is a p-value?

A number produced by the data after the study; it differs each time you sample.

What is the key difference between alpha and the p-value?

Alpha is set in advance, whereas the p-value is calculated from the study data.

What is the decision rule when p < α?

Reject H₀; the result is statistically significant.

What is the decision rule when p > α?

Fail to reject H₀; the result is not statistically significant.

If alpha and the p-value have the same numerical value, are they the same thing?

No. They can share the same number by coincidence, but alpha is a preset threshold and the p-value comes from the data.

What was the baseline sample-size example in the lecture?

Comparing two group means with an expected difference of 5, SD = 10, α = 0.05, and power = 80%.

What sample size was required in the baseline example?

Approximately 63 patients per group.

What happens to the required sample size when alpha decreases from 0.05 to 0.01 in the lecture example?

It increases from approximately 63 to 94 patients per group.

Why does a stricter alpha threshold require more data in the lecture example?

A stricter significance threshold creates a higher bar for detecting an effect.

What happens to the required sample size when power increases from 80% to 90% in the lecture example?

It increases to approximately 85 patients per group.

Why does increasing power require more data in the lecture example?

Being more certain that a real effect will be detected requires a larger sample; lowering beta costs data.

What happens to sample size when the expected effect size doubles from 5 to 10 in the lecture example?

The required sample size drops to approximately 16 patients per group at α = 0.05 and power = 80%.

Why does a larger effect size require fewer patients to detect in the lecture example?

A larger difference is easier to detect.

What happens to the sampling distribution of the mean as sample size increases?

It becomes more normal-shaped and narrower, with a bigger peak in the middle.

What is the formula for standard error (SE) in the Central Limit Theorem slide?

SE = SD / √n.

What does SD represent in the standard error formula?

Standard deviation.

What does n represent in the standard error formula?

Sample size.

Why does standard error decrease as sample size increases?

The square root of sample size is in the denominator, so increasing n shrinks the spread.

Why does averaging reduce noise in a larger sample?

Unusually high and unusually low values are more likely to balance each other out.

How does the Central Limit Theorem relate to confidence intervals?

A larger sample size produces a narrower confidence interval because the sampling distribution becomes narrower.

What is the overall purpose of quantitative research inference?

To use sample statistics to make inferences about population parameters.

What is the first step in the general quantitative research process shown in the lecture?

Review the literature and formulate hypotheses as part of proposing research questions.

What comes after proposing research questions in the quantitative research process?

Select a sample.

What topics are associated with selecting a sample in the quantitative research process?

Sampling methods, sampling error, sampling bias, sample representativeness, and determining sample size.

What comes after selecting a sample in the quantitative research process?

Collect data.

What concepts are considered during data collection?

Types of variables and levels of measurement, and measurement reliability and validity.

What comes after collecting data in the quantitative research process?

Present data and analyze data.

What are the methods of data presentation listed in the lecture?

Textual, tabular, and graphical presentation.

What descriptive statistics topics are listed in the quantitative research process?

Frequency distribution, measures of central tendency, and measures of variability.

What topics are included under inferential statistics in the lecture?

Probability, the Central Limit Theorem, hypothesis testing, and statistical significance.

What is another consideration when analyzing quantitative research data?

Deciding which statistical tests to use based on the level of measurement.

What is random error?

Natural random noise between a sample statistic and the true population value.

How does random error behave across samples?

It varies from sample to sample.

What happens to random error as sample size increases?

It shrinks.

What do standard error and margin of error measure?

Random variation or error associated with estimates, as discussed in the confidence interval material.

What is bias in research?

A systematic, one-directional distortion built into how a study is designed or conducted.

Does bias average out with a larger sample size?

No. Bias does not average out simply by increasing sample size.

What is the difference between random error and bias?

Random error varies across samples and shrinks with larger samples; bias systematically distorts results and does not disappear just by increasing sample size.

Can a large study still be badly biased?

Yes. A larger sample fixes error, not bias.

What is an example of sampling-frame bias from the lecture?

Using a flawed sampling frame.

What is an example of measurement bias involving blood pressure?

A blood pressure cuff that consistently reads 2 mmHg too high.

What is sampling bias?

A systematic difference between the people who end up in a sample and those in the target population.

What is self-selection bias?

Bias that occurs when participants choose to opt themselves into a study.

What is an example of self-selection bias from the lecture?

Only students with strong opinions complete the SPOT evaluations.

What is non-response bias?

Bias that occurs when people who do not respond differ systematically from those who do respond.

What is survivorship bias?

Bias that occurs when only the survivors or successful cases remain visible.

Why is sampling bias only one source of bias?

Bias can enter research at almost any stage, not just when participants are selected.

What is measurement (information) bias?

Systematic inaccuracy in how data are collected or recorded.

What is an example of measurement bias involving charting?

Inconsistent charting between nurses.

What is recall bias?

Bias that occurs when participants with a certain outcome remember past exposures differently from participants with another outcome.

What is an example of recall bias involving medication?

Patients with a poor outcome remember a missed medication dose more vividly than other patients.

What is observer bias?

Bias in which a researcher’s expectations unconsciously shape what they record or how they interpret results.

What is confounding?

A situation in which an unmeasured third variable is related to both the exposure and outcome, creating an association that is not actually causal.

What is publication bias?

Bias in which studies with statistically significant or positive findings are more likely to be published, skewing the overall evidence base.

A wound-care study interviews only patients who still attend follow-up. What bias may occur?

Survivorship bias, because patients who dropped out, possibly due to treatment failure, are missing.

A nurse evaluates the usability of a charting tool she developed. What bias may occur?

Observer bias.

A survey asks, “How much did you enjoy our excellent care?” What bias may occur?

Measurement bias due to a leading question.

A meta-analysis includes only published trials and may overestimate a treatment’s effect. What bias may occur?

Publication bias.

What question should researchers ask before trusting a result?

Who or what could be missing or distorted, and in which direction?

A pain-management study surveys only patients who complete a six-week follow-up, excluding those who drop out because of unmanaged pain. What bias is this?

Survivorship bias.

A nurse who believes a turning protocol prevents pressure injuries assesses patients’ skin for injuries. What bias is this?

Observer bias.

A burnout survey is emailed to all staff, but only the most frustrated nurses respond. What bias does this illustrate?

Self-selection bias.

A wound-dressing review includes only published studies in which the dressing worked well. What bias does this illustrate?

Publication bias.

Patients with a bad outcome recall skipping a medication dose more vividly than patients who did well. What bias does this illustrate?

Recall bias.

Units with more nurse overtime have worse outcomes, but understaffing may drive both. What bias does this illustrate?

Confounding.

A manual blood pressure cuff has not been calibrated in two years and consistently reads high. What bias does this illustrate?

Measurement (information) bias.

What is the null hypothesis (H₀)?

The default, skeptical position that there is no difference or no effect.

What is the alternative hypothesis (H₁)?

The hypothesis that a difference exists; what the researcher expects or hopes to find.

What does a p-value represent in the lecture’s hypothesis-testing preview?

The probability that the result happened by chance alone, if H₀ were true.

What alpha value is almost always used as the threshold for statistical significance in the lecture?

0.05.

What does p < 0.05 mean when α = 0.05?

Reject H₀; the result is statistically significant.

What does p > 0.05 mean when α = 0.05?

Fail to reject H₀; the result is not statistically significant.

What is the relationship between confidence intervals and hypothesis testing?

Confidence intervals provide another way to express the same general idea of statistical significance while showing uncertainty around an estimate.

What is a population parameter?

A numerical characteristic of a population that researchers want to estimate or draw inferences about.

What is a sample statistic?

A numerical value calculated from a sample and used to make inferences about a population parameter.

How do sample statistics relate to population parameters?

Researchers use sample statistics to make inferences about population parameters.

What kind of sample is needed for population inference in the lecture’s confidence interval diagram?

A representative sample with a sufficient sample size.

Why should inferences include a probability statement?

To reflect uncertainty caused by random variation and sampling error.

In the blood pressure drug example, what does H₀ state?

There is no difference in blood pressure between the treatment and control groups.