1: Measurement and sampling (Key Odd)

Review Questions

  1. Statistics serves many different scientific disciplines, helping to make its observations and results more informative and reliable.
  2. Measurement error is the difference between what appears on a data collection form and the true state of affairs. Processing error refers to errors that occur during data handling (e.g., during data entry).
  3. GIGO = "Garbage in, garbage out" 
  4. Definitions -- see online notes
  5. Statistics is not merely a compilation of computational techniques. It is a way of learning from data that involves gaining knowledge from data. It is also about collecting and organizing data. Good statisticians avoid mechanical interpretations of numerical results. The focus on uncovering the true nature of things through careful study design and analysis. 
  6. Categorical measurement =  nominal measurement = qualitative measurement; Quantitative  variable =  scale variable 
  7. Observations in rows, variables in columns, and values in cells. 
  8. Sampling with replacement means that a once a unit is sampled from the population they are tossed back in and may be sampled again. Sampling without replacement means that the sample unit is ineligible for a second selection. In most studies, the sampling fraction is small enough so as to make the distinction between sampling with or without replacement is inconsequential. However, when the sampling fraction get large (e.g., 5 - 10%), then a distinction must be made between sampling with and without replacement. Most elementary statistical techniques assume that sampling is done with replacement or that the sampling fraction is small enough, making the distinction inconsequential.
  9. The sampling fraction
  10. (c)
  11. (b)
  12. (a) equal  . .  . (b) does not
  13. This sample is not a random because it excludes people toward the end of the alphabet. Simple random samples are selected so that everyone in the population has an equal chance of selection and every conceivable sample is possible. 

Exercises

1.1 Cargo cult science. Brief essay. No key provided.

1.3 Hospital duration data (HDUR). Population = all patients who were hospitalized in this hospital during the period of the study. Sample = the 30 patients that ended-up in this study.

1.5 Teaching effectiveness. Student ratings are subjective measures of satisfaction. A more meaningful measure of teaching effectiveness could be defined  in terms of improving students' abilities to reason and solve meaningful problems. The better the improvement in student problem solving (on standardized tests), the more effective the teacher. 

 

1.7 Hospital duration data (HDUR).

(A) The following variables are categorical : SEX, AB, CULT, SERV. The following variables are quantitative : DUR, AGE, TEMP, WBC. There are no ordinal variables in this data set.

(B) The value of DUR for observation 4 is 11.

(C) The value of AGE for observation 4 is 43.

1.9 Variable types 1 (A) ordinal (B) categorical  (C) quantitative (D) quantitative (E) quantitative (F) quantitative (G) categorical  (H) categorical