Page 32

navidi_monk_essential_statistics_1e_ch1_3

30 Chapter 1 Basic Ideas Chapter 1 Summary Section 1.1: Most populations are too large to allow us to study each member, so we draw samples and study those. Samples must be drawn by an appropriate method. Simple random sampling, stratified sampling, cluster sampling, and systematic sampling are all valid methods. When none of these methods are feasible, a sample of convenience may be used, so long as it is reasonable to believe that there is no systematic difference between the sample and the population. Section 1.2: Data sets contain values of variables. Qualitative variables place items in categories, whereas quantitative variables are counts or measurements. Qualitative variables can be either ordinal or nominal. An ordinal variable is one for which the categories have a natural ordering. For nominal variables, the categories have no natural ordering. Quantitative variables can be discrete or continuous. Discrete variables are ones whose possible values can be listed, whereas continuous variables can take on values anywhere within an interval. Section 1.3: Scientists conduct studies to determine whether different treatments produce different outcomes. The most reliable studies are randomized experiments, in which subjects are assigned to treatments at random. When randomized experiments are not feasible, observational studies may be performed. Results of observational studies may be hard to interpret, because of the potential for confounding. Section 1.4: Some studies produce more reliable results than others. A study that is conducted by a method that tends to produce an incorrect result is said to be biased. Some of the most common forms of bias are voluntary response bias, self-interest bias, social acceptability bias, leading question bias, nonresponse bias, and sampling bias. Vocabulary and Notation bias 26 leading question bias 27 sample of convenience 5 biased 27 nominal variable 13 sampling bias 28 case-control study 23 nonresponders 28 seed 4 categorical variable 13 nonresponse bias 28 self-interest bias 27 cluster sample 6 observational study 19 simple random sample 3 cohort study 23 ordinal variable 13 social acceptability bias 27 completely randomized experiment 21 outcome 19 statistic 8 confounder 21 parameter 8 statistics 2 confounding 21 population 2 strata 5 continuous variable 14 prospective study 23 stratified sample 5 cross-sectional study 23 qualitative variable 13 subject 18 data 12 quantitative variable 13 systematic sample 6 data set 12 randomized block experiment 21 treatment 19 discrete variable 14 randomized experiment 19 unbiased 27 double-blind 20 response 19 variable 12 experimental unit 18 retrospective study 23 voluntary response bias 27 individual 12 sample 2 voluntary response sample 7 Chapter Quiz 1. Provide an example of a qualitative variable and an example of a quantitative variable. 2. Is the name of your favorite author a qualitative variable or a quantitative variable? 3. True or false: Nominal variables do not have a natural ordering. 4. variables are quantitative variables that can take on any value in some interval. 5. True or false: Ideally, a sample should represent the population as little as possible. 6. A utility company sends surveys to 200 of its customers in such a way that 100 surveys are sent to customers who pay their bills on time, 50 surveys are sent to customers whose bills are less than 30 days late, and 50 surveys are sent to customers whose bills are more than 30 days late. Which type of sample does this represent? 7. A sample of convenience is when it is reasonable to believe that there is no systematic difference between the sample and the population. (Choices: acceptable, not acceptable) 8. The manager of a restaurant walks around and asks selected customers about the service they have received. Which type of sample does this represent? 9. True or false: An experiment where neither the investigators nor the subjects know who has been assigned to which treatment is called a double-blind experiment. 10. A poll is conducted of 3500 households close to major national airports, and another 2000 that are not close to an airport, in order to study whether living in a noisier environment results in health effects. Is this a randomized experiment or an observational study?


navidi_monk_essential_statistics_1e_ch1_3
To see the actual publication please follow the link above