All Free Statistics MCQs with Answers
Every Statistics question in the bank, across all chapters, each with the correct answer and a written explanation. Free and unlimited, with no account needed.
545 questions · page 29 of 55
281. In inferential statistics, we study____________?
- A. the methods to make decisions about population based on sample results
- B. how to make decisions about mean, median, or mode
- C. how a sample is obtained from a population
- D. None of the above
Explanation: Inferential statistics uses sample results to draw conclusions or make decisions about the wider population. Describing a sample or explaining how it was selected does not by itself constitute inference.
Correct answer: the methods to make decisions about population based on sample results282. A numerical value calculated from sample is called____________?
- A. Mathematics
- B. Economics
- C. Statistic
- D. None of these
Explanation: A numerical measure calculated from a sample is called a statistic, such as the sample mean or sample proportion. The corresponding numerical measure for a population is called a parameter.
Correct answer: Statistic283. We use non parametric tests when the data is_________?
- A. Qualitative
- B. Ranked
- C. Both a and b
- D. None of these
Explanation: Non-parametric methods are suitable for qualitative data and for ordinal or ranked observations because they do not require the usual population distribution assumptions. Thus both listed types of data can lead to non-parametric testing.
Correct answer: Both a and b284. An alternative non parametric test of ANOVA?
- A. Sign test
- B. Wilcoxon sign rank test
- C. Median test
- D. Kruskal Wallis test
Explanation: The Kruskal-Wallis test is the rank-based non-parametric counterpart of one-way ANOVA for comparing several independent groups. The sign and Wilcoxon tests are mainly used for one- or two-sample situations.
Correct answer: Kruskal Wallis test285. When the assumption of normality not met we go for__________?
- A. Parametric test
- B. Non parametric tests
- C. Either both
- D. None of these
Explanation: When the normality assumption is seriously violated, a non-parametric test is often preferred because it does not require a normal population distribution. Parametric tests may still be usable when their robustness conditions are adequate.
Correct answer: Non parametric tests286. When in hypothesis testing (alpha) increases (1-beta)?
- A. Decrease
- B. Equal
- C. Increases
- D. None of these
Explanation: With the sample size and test design fixed, increasing alpha generally makes rejection easier, which increases power, 1-beta. This illustrates the usual trade-off between Type I error and Type II error.
Correct answer: Increases287. In hypothesis testing probability of (alpha +beta) =1?
- A. Increase
- B. Decrease
- C. Yes
- D. No
Explanation: Alpha and beta are error probabilities under different hypotheses, so they are not complements in general. Consequently, their sum need not equal 1.
Correct answer: No288. The non parametric equivalent of an unpaired samples t-test____________?
- A. Kruskal Wallis Test
- B. Mann-Whitney U test
- C. Wilcoxon signed rank test
- D. Sign test
Explanation: The Mann-Whitney U test is the usual nonparametric alternative to the independent or unpaired two-sample t-test. The Wilcoxon signed-rank test is instead used for paired samples, while Kruskal-Wallis handles more than two independent groups.
Correct answer: Mann-Whitney U test289. When testing for randomness, we can use_____________?
- A. Sign test
- B. None of these
- C. Runs test
- D. D Mann-Whitney U test
Explanation: The runs test examines whether the order of observations is random by comparing the observed number of runs with what randomness would produce. Sign, Mann-Whitney, and related tests assess other distributional or location questions.
Correct answer: Runs test290. Which of the following test is most likely assessing this null hypothesis: Ho The number of violations per apartment in the population of all city apartments is binomially distributed with a probability of success in any one trial of P=0.3 dd ?
- A. The Mann-Whitney test
- B. The Kruskal-Wallis test
- C. The Wilcoxon Signed Rank test
- D. The Kolmogorove-Simirnor test
Explanation: This is a goodness-of-fit question asking whether observed counts follow a specified binomial distribution. The Kolmogorov-Smirnov test compares an empirical distribution with a hypothesized distribution, making option d the intended choice.
Correct answer: The Kolmogorove-Simirnor test