EXAM 1 STATISTICS

T or F: More types of calculations can be performed with data at the nominal level than with data at the interval level.

False. Less types of calculations can be performed with data at the nominal level than with data at the interval level. More types of calculations can be performed with data at the interval or ratio level than at the nominal level.

Name each level of measurement for which data can be qualitative.

Nominal, Ordinal.

Open Question: What can be done to get students to eat healthier foods? Closed Questions: How would you get students to eat healthier foods? Advantage and Disadvantage for both types of questions

Open Question Advantage: Can give any response Disadvantage: Hard to analyze and compare Closed Question Advantage: Easy to analyze Disadvantage: Your answer may not be there but you still have to choose one.

Name each level of measurement for which data can be quantitative.

Ordinal, Interval, and Ratio.

Describe two methods you can use to generate random numbers.

Physical mixing (names in a hat), Table of random numbers, Computer random digit dialing

Population or Sample? The age of each member of the house of representatives.

Population. It includes all the members of the house of representatives.

Population vs the Sample: Ages of adults in the United States who own computers. Ages of adults in the Unites States who own Dell Computers.

Population: Ages of adults who own computers. Sample: Ages of adults who own dell computers.

Population vs the Sample: A survey of 1,045 registered voters found that 19% think the economy is the most important issue to consider when voting for congress.

Population: All registered voters in the United States. Sample: 1,045 registered voters. The 19% is a value from the sample so it is a statistic.

Qualitative or Quantitative? Data sets level of Measurement? The region representing the top salesperson in a corporation for the past six years is given. Southeast Northwest Northeast Southeast Southwest Southwest

Qualitative and Nominal. Categories only.

Qualitative or Quantitative? Data sets level of measurement? The top 5 football teams in the final college football poll. 1. Florida 2. Ohio State 3. LSU 4. USC 5. Boise State

Qualitative and Ordinal. They are categories that you can put in order but their is no meaningful numerical difference between the categories, just a ranking.

Qualitative or Quantitative? The telephone numbers in a telephone directory?

Qualitative, they are numbers but they are not numbers you can do math with, nor are they numbers that you can put in a numerical order.

Qualitative or Quantitative? Data sets level of Measurement? The lengths (in inches) of a sample of striped bass caught in Maryland waters are listed. 16 17.25 19 18.75 21 20.3 19.8 24 21.82

Quantitative and Ratio (has inherent zero).

Population or Sample? A survey of 500 spectators from a stadium with 42,000 spectators.

Sample. It is a survey of only 500 spectators when there is a total of 42,000. 42,000 is the population.

Identify the sampling technique used and discuss potential sources of bias: Using a random digit dialing, researchers called 1599 people and asked what obstacles (such as childcare) kept them from exercising.

Simple Random Sampling because the people all had the same chance of being selected. The bias might be that people without phones cannot be selected.

Identify the sampling technique used and discuss potential sources of bias: Chosen at random, 1210 hospital outpatients were contacted and asked their opinion of the care they received.

Simple random sample because each outpatient had an equal chance of being chosen. It does not matter that they are only using outpatients because that is probably the only patients they want information about. No obvious bias.

What is the Descriptive branch of Statistics here? A Survey of 546 women found that more than 56% are the primary investor in their household.

The 56% that are the primary investor in their household is the descriptive statistic. Inferential Stats are the conclusions so anything that shows a relationship between the sample and the whole population. Ex: 56% of all household primary investors are women.

Television Program ratings by Nielsen Media Research are described on page 17. Discuss the Strata used in the example.

The Strata (groups) used in the example are locations, ethnic groups and income brackets.

What are descriptive statistics?

The graphs, averages, percentiles etc are the descriptive stats.

A simulation

the use of a mathematical or physical model to reproduce the conditions of a situation or process.

Performing an Experiment

A treatment is applied to part of the population and responses are observed. Another part of the population may be used as a control group.

Identify the sampling technique used and discuss potential sources of bias: A list of managers is compiled and ordered. After a starting number is randomly chosen, every ninth name is selected until 1000 managers are selected. The managers are questioned about the use of digital media.

Systematic sampling because every ninth name on the list is selected. Bias not obvious.

Observational Study

a researcher observes and measures characteristics of interest of part of a population but does not change existing conditions.

T or F: Using a systematic sample guarantees that members of each group within a population will be sampled.

False, Using a stratified sample guarantees that members of each group within a population will be sampled.

How is a Sample related to a population?

A sample is a subset of a population. It is important that a sample be representative of a population to reduce bias.

Method of Data collection A study of how fast a virus would spread in a metropolitan area.

A simulation. One would not want to release an actual virus so you could use math or a computer to simulate it.

T or F: Data at the ordinal level are quantitative only.

False. Data at the ordinal level can be qualitative or quantitative.

T or F: For Data at the interval level, you cannot calculate meaningful differences between data entries.

False. For data at the interval level you can calculate meaningful differences between data entries. At the Nominal or Ordinal level you cannot calculate meaningful differences.

What is the difference between a parameter and a statistic?

A PARAMETER is a numerical (data) description of a population. A STATISTIC is a numerical description of a Sample.

The following appear on a physicians intake form. Identify the level of measurement of the data. a. Temperature b. Allergies c. Weight d. Pain Level (0 to 10)

A. Quantitative and Interval B. Qualitative and Nominal C. Quantitative and Ratio D. Ordinal.

A pharmaceutical company wants to test the effectiveness of a new allergy drug. The company identifies 250 females age 30-35 years old who suffer from severe allergies. One group has new medicine and the other has a placebo pill. After 6 months the subjects symptoms are studied. A. Identify the experimental units in this experiments B. How many treatments are used in this experiment? C. Identify a potential problem with the experimental drug being used and suggest a way to improve it. D. How could this experiment be designed to double blind?

A. The 250 females who have allergies are the experimental units B. 2, the new drug and the placebo pill C. Comparing new medicine to a placebo pill might not show how good the medicine is compared to other medicines. I would recommend that they give one group the new medicine and the other group a different medicine that has been used for years. D. By making sure the neither the subjects nor the experimenter know which group got which treatment.

Identify the sampling technique used and discuss potential sources of bias: Questioning students as they left a university library, a researcher asked 358 students about their drinking habits.

Convenience sampling because the researcher is just standing outside the library and using a sample that is readily available at that location. A potential bias is that the researcher is only sampling people who use the library, leaving out students who do not attend the library who may be less studious.

Biased or Not? Why is drinking fruit juice good for you?

Biased because it assumes fruit juice is good for you. Better wording would be "Is drinking Fruit juice good or not good for you?" (Gives both sides)

Census or Sampling: The average salary of 50 employees of a company.

Census. Can easily get the salaries of 50 employees.

Method of Data collection A study of the effect on the human digestive system of potato chips made with a fat substitute.

Experiment. You can perform this experiment without causing harm to the digestive system

T or F: A statistic is a measure that describes a population characteristic.

False! A statistic is a measure that describes a sample characteristic. A Parameter describes a population Characteristic.

T or F: A population is the collection of some outcomes, responses, measurements, or counts that are of interest.

False, a population is the collection of all outcomes, responses, measurements, or counts that are of interest. A sample is the collection of some outcomes, responses, measurements, or counts that are of interest.

T or F The method for selecting a stratified sample is to order a population in some way and then select members of the population at regular intervals.

False, using a systematic sample is to order a population in some way and then select members of the population at regular intervals.

Identify the sampling technique used and discuss potential sources of bias: Soybeans are planted on a 48 acre field. The field is divided into one acre subplots. A sample of plants is taken from each subplot to estimate the harvest.

Stratified Sampling because groups are created based on acreage and a sample is taken from each subplot. Type of soil or plot location may create bias otherwise, no bias.

T or F In a randomized block design, subjects with similar characteristics are divided into blocks, and then within each block, randomly assigned to treatment groups.

True.

T or F: It is Impossible for the Census Bureau to obtain all the census data about the population of the United States.

True.

Biased or Not? How many hours of sleep do you get on an average night?

Unbiased. But professor does not like the word average. Also could question the word "night" because not everyone sleeps at night.

What is the difference between an observational study and an experiment?

With an observational study one just observes and takes notes (measurements) but does not manipulate the units. With an experiment one does manipulate the units with treatments and controls and then takes measurements.

A survey

an investigation of one or more characteristics of a population (usually carried out on people by asking questions)