Key Concepts in Statistical Sampling and Analysis

Key Concepts in Statistical Sampling and Analysis

Sampling and Population

1. Which of the following is not an advantage of using a sample versus a census?
   Ready access to respondents.
2. Which of the following refers to the basic level of investigation?
   A sample unit.
3. Which of the following describes an entire group under study as specified by the objectives of the research project?
   Population.
4. Of the following, which is not true of nonprobability samples?
   Can compute the chances of any one population member being selected.
5. With non-probability methods, there is no way to determine the probability even if the population size is known because…
   Selection technique is subjective.

Sample Size Determination

6. Which of the following is not one of the three ways a researcher may estimate variability?
   By examining research objectives against costs.
7. _________ is the most correct method of determining a sample size.
   The confidence interval approach.
8. ____ is reflected in the spread of the distribution.
   Amount of variability.
9. An accounting of the complete population is a _____.
   Census.

Hypothesis Testing and Statistical Analysis

10. ____ testing is the comparison of the sample statistic with a population value proposed prior to undertaking the study.
    Hypothesis.
11. Why do most computer statistical programs report only the t value?
    Because it is identical to the z value with large samples.
12. The ___ is the variable customarily termed y in the regression straight-line equation.
    Dependent variable.
13. Statistical tests are used when a researcher wants to ____ of two different groups or samples.
    Compare the means or percentages.
14. The flipping of a coin to decide heads or tails or lottery numbers are examples of what type of random sampling?
    Random device method.
15. The time and effort expended in the research process are _____
    Wasted if the report does not communicate effectively.
16. Which of the following is not one of the three values used in parameter estimation?
    The desired population.
17. ANOVA is a ____ procedure that signals when at least one pair of means is significantly different.
    Green light.
18. ____ is the term indicating the presence of correlations among the independent variables in multiple regressions.
    Multicollinearity.
19. Running ____ of the significance of the difference between means is much less advantageous than using ANOVA.
    Multiple t tests.

Scales and Measurement

A respondent is asked to indicate his or her first, second, third, and fourth choices of hotels. This means the choices are __ scaled.
Ordinally scaled. Stapel scale relies not on bipolar terms, but on positive and negative numbers, typically ranging from +5 to -5, but sometimes from +3 to -3, depending on the discrimination abilities of respondents. Symmetric interval scale is “balanced”, as it has equal amounts of positive and negative positions, and typically it has “no opinion” or “neutral” separating the negative and positive sides. Any self-administered survey qualifies for anonymity as long as the respondent does not indicate his or her identity and provided the questionnaire does not have any covert identification tracing mechanism.
TRUE. An accounting of the complete population is a CENSUS.

Advanced Sampling and Analysis

Using computer-generated numbers is a tractable and more sophisticated application of: simple random sampling. The most correct method of determining sample size is the confidence interval approach, which applies the concepts of all of the following except: Validity and Value. All errors in a survey except those attributable to the sample plan and sample size are termed: Non-sampling Error. A researcher who wants to see if variables, such as advertising recall scores, positively associate with intentions to buy the advertised brand would use: Association. Which of the following is not one of the measures of variability, with each measure providing its own unique version of information that helps to describe the diversity of response?
Mode. Researchers often rely on the standard deviation when performing basic analyses, and usually report it in their tables: FALSE.

Interpreting Statistical Findings

Differences are important, but a finding of statistical significance in no way guarantees a Meaningful difference. Even though they are not cause-and-effect relationships, which of the following often provide researchers with insights that lead to understanding?
Statistical linkages or relationships. The ___ is the variable customarily termed Y in the regression straight-line equation.
Dependent. Presentation of table after table of statistics by researchers Has little value unless tied to an explanation of why data are important. Which of the following is an advantage of reporting all important limitations?
Actually add credibility to the research.