Erin - Online Resources



DISCUSSION GROUP ANSWERS

CHAPTER 2: UNDERSTANDING DATA DISTRIBUTIONS WITH TABLES AND GRAPHS

1. What is the level of measurement for each of the following variables? Why?

a. Length of incarceration in months (starting at 0 meaning no offenses)

a. Ratio—nonarbitrary zero

b.

b. Number of self-reported deviant acts (measured in categories of 0–4, 5–9, 10–14, and 15 or more)

a. Ordinal—rank ordered data

b.

c. Types of violent crimes (measured in categories of “aggravated assault,” “rape,”

a. “robbery,” and “homicide”)

b. Nominal—because it is categorical

2. In a study examining the effect of child abuse on subsequent juvenile delinquency, what is the independent variable and what is the dependent variable?

            Independent variable (IV): Child abuse

            Dependent variable (DV): Subsequent juvenile delinquency

3. In 2001, Maryland had the highest robbery rate of any state in the country. If its population was 5,575,156 people, and there were 13,125 total robberies, what was the robbery rate for the state per 100,000 residents in the state?

                (13,125 / 5,575,156) * 100,000 = 235.42 per 100,000

4. In 2001, there were 3,840 robbery offenses and 208 rape offenses known to the police in the city of Washington, D.C. In that year, what was the ratio of robberies to rapes?

               3,840 to 208 = 3,840 /208 =18.46 robberies to each rape

For the following questions (5–8), remember that the independent variable is the variable thought to affect or in some way contribute to fluctuations or changes in the dependent variable.

5. Suppose that you as a researcher wanted to examine the impact of police officer rank on perceptions of friendliness by citizens in College Park, Maryland. Rank is measured as patrol officer, sergeant, or captain, and citizen perception is captured by a scale ranging from -50 to 50 and measured in a survey given to 1,000 randomly selected people after an interaction with a police officer. Five hundred people talked to a patrol officer, 300 talked to a sergeant, and 200 people talked to a captain.

a. Identify the independent and dependent variables.

            IV: Officer rank

            DV: Citizens' perceptions of friendliness

b. For each variable, name what kind of variable it is (nominal, ordinal, interval, or ratio).

            IV: Ordinal

            DV: Interval

6. A researcher is interested in the relationship between attention-deficit/hyperactivity disorder (ADHD) and conduct disorder in middle school classes. Her hypothesis is that students with ADHD are more likely to experience conduct disorders at school. She obtained a random sample of seventh and eighth graders from a suburban middle school in Washington, D.C. ADHD was measured using a self-administered test that ranked the student on the following order: no attention deficit, some attention deficit, high attention deficit; and conduct disorder was a dichotomous (i.e., two category) variable consisting of conduct disorder or no conduct disorder.

a. What are the measurement unit and population for this sample?     

 Unit of Observation: Students         

Population: Seventh- and eighth-grade suburban middle school students in Washington, D.C.

b. Identify the independent and dependent variable.

            IV: ADHD

            DV: Conduct disorder

c. For each variable, name what kind of variable it is (nominal, ordinal, interval, or ratio).

            IV: Ordinal

            DV: Nominal

7. A researcher is interested in whether a rehabilitation program reduces how often prisoners will reoffend after they are released. Two hundred inmates at Maryland Correctional Institution in Jessup are randomly assigned to participate in the program while another 200 inmates are kept in prison without participating in any rehabilitative programs. All of the inmates are then studied for two years after their release to see how many times they were arrested again.

a. What are the measurement unit and population for this sample?

           Unit of Observation: Inmates

            Population: Inmates at the MD Correctional Institution in Jessup

b. Identify the independent and dependent variable.

            IV: Rehabilitation program participation

            DV: Rearrest upon release from prison

c. For each variable, name what kind of variable it is (nominal, ordinal, interval, or ratio).

            IV: Nominal

            DV: Ratio

8. The Women’s Resource Center of Maryland (WRCM) has just hired you as a statistical consultant. Your first task is to investigate whether or not rape reporting legislation (measured as a dichotomy: rape legislation; no rape legislation) affects the reporting rate for sexual assaults in Maryland counties (measured as the number of rapes per 100,000 residents). You take a random sample of four Maryland counties, two of which have the legislation (Baltimore and Frederick) and two of which do not (Prince Georges and Montgomery). Below are your data:

|Rape Rate per 100,000 Residents in a Sample of Maryland Counties |

|County |# of Rapes Reported |County Population |Rape Reporting Rate |

|  |  |  |  |

|Prince George |517 |533,398 |96.9 |

|Montgomery |371 |422,384 |87.8 |

|Baltimore |2,124 |1,042,350 |203.8 |

|Frederick |397 |209,751 |189.3 |

a. Calculate the rape rate per 100,000 residents for each county in the table above.

b. What are the measurement unit and population for this sample?

a. Unit of Observation: County

b. Population: All MD Counties

c. Identify the independent and dependent variable and name the level of measurement for each (nominal, ordinal, interval, or ratio).

a. IV: Rape reporting legislation (nominal)

b. DV: Rape reporting rate (ratio)

d. 9. The following table presents data from the Uniform Crime Reports about crime reported to the police in Maryland from 1998–2002. Using these data, answer the following questions:

|Year |# Index Offenses |a) Index Offense Rate |Population |# Violent Offenses |b) Violent Offense Rate|# Property |

| | |(per 100,000) | | |(per 100,000) |Offenses |

|1998 |250,654 |4959.42 |5,054,100 |36,137 |715.00 |214,517 |

|1999 |254,420 |4919.18 |5,172,000 |38,447 |743.37 |215,973 |

|2000 |259,635 |4930.09 |5,266,330 |40,102 |761.48 |219,533 |

|2001 |261,600 |4866.84 |5,375,156 |42,088 |783.01 |219,512 |

|2002 |263,950 |4797.43 |5,501,900 |44,385 |806.72 |219,565 |

a. For each year (1998–2002), calculate the index offense rate (per 100,000 population). What do these values mean?

The Number of Index Crimes for every 100,000 population

b. For each year (1998–2002), calculate the violent offense rate (per 100,000 population).

c. What is the ratio of the 2000 property offense rate to the 2002 property offense rate? (Hint: You will need to calculate these rates before creating your ratio). What does this ratio mean?

4168.62 / 3990.71 = 1.04 For every property crime committed in 2002 there were 1.04 committed in 2000.

d. In 1999, what proportion of index offenses were property offenses?

215973 / 254420 = .849

e. In 2000, what percentage of index offenses were violent offenses?

40102/259635 = .154 .154*100 = 15.4%

f. Graph your calculated violent offense rate by year, and explain why you chose the type of graph you did to represent your data.

Data over time is generally graphed with a line graph known as a “Time Plot.”

[pic]

g. What type of data are offenses and rates (qualitative or quantitative; and nominal, ordinal, interval, or ratio)?

Offense = qualitative, ordinal; Rate = quantitative, ratio

10. After it was ruled to be unconstitutional, the death penalty was reinstituted in 1976. Since then, 548 individuals have been executed. Of those executions, 391 were by lethal injection, 141 were by electrocution, 11 were by gas chamber, 3 were by hanging, and 2 were by firing squad.

a. Execution type is at what level of measurement?

Nominal

b. What proportion of executions involved electrocution?

141 / 548 = .257

c. What percent of executions involved the firing squad?

(2 / 548) * 100 = 0.37%

d. Create a graph to illustrate the distribution of these data.

[pic]

11. You are interested in analyzing crime on college campuses, so you collect information about the number of crimes reported to the University of Maryland campus police and the number of arrests made during 2003. The campus police report that they made 114 arrests for liquor law violations, 72 arrests for drug offenses, and 16 arrests for weapons violations.

a. What proportion of arrests made by the campus police were for drug offenses?

72 / 202 = .356

b. What percentage of arrests were for weapons violations?

16 / 202 = .079 x 100 = 7.9%

c. Create a pie chart to illustrate the distribution of arrests for liquor law violations, drug offenses, and weapons violations.

[pic]

In the same analysis, you collect the following data on the number of crimes reported to the campus police during 2003.

|Type of Offense |Frequency |Prop |% |

|Murder |0 |0 |0 |

|Rape |5 |0.006 |0.6% |

|Robbery |13 |0.015 |1.5% |

|Aggravated Assault |13 |0.015 |1.5% |

|Burglary |107 |0.119 |11.9% |

|Larceny |725 |0.809 |80.9% |

|Motor Vehicle Theft |33 |0.037 |3.7% |

|Total |896 |1 |100% |

d. What proportion of crimes reported were aggravated assaults?

13 / 896 = .015

e. What percentage of crimes reported were larcenies?

725 / 896 = .809 x 100 = 80.9%

f. Create a percentage bar graph to illustrate the distribution of offenses.

[pic]

12. The following hypothetical data represent the number of housing inspections made to each of 30 public housing complexes in Montgomery County. With this data, construct an ungrouped frequency distribution. Calculate the frequency, percentage, cumulative frequency, and cumulative percent. Report your table and then use it to answer the following questions:

|1 |0 |2 |3 |4 |

|0 |1 |0 |2 |3 |

|6 |0 |1 |0 |2 |

|1 |7 |0 |1 |0 |

|0 |2 |2 |0 |1 |

|1 |3 |3 |3 |0 |

Answer:

|# Visits |Complexes |cf |% |Cumulative % |

|0 |10 |10 |33.33% |33.33% |

|1 |7 |17 |23.33% |56.66% |

|2 |5 |22 |16.67% |73.33% |

|3 |5 |27 |16.67% |90.00% |

|4 |1 |28 |3.33% |93.33% |

|5 |0 |28 |0.00% |93.33% |

|6 |1 |29 |3.33% |96.66% |

|7 |1 |30 |3.33% |99.99% |

Note: Percentages do not sum to 100% due to rounding.

a. What percentage had less than 3 visits?

73.33% had less than 3 visits.

b. What percentage had exactly 3 visits?

16.67% had exactly 3 visits.

c. What percentage had more than 3 visits?

9.99% had more than 3 visits.

d. Construct a frequency histogram with this data.

[pic]

13. We collected the property crime rates for a random sample of 85 U.S. cities. With the following grouped frequency distribution, answer the questions below:

Property Crime Rates (per 100,000)

Stated Limits Real Limits Mid f % cf c%

0–99 -.5 –99.5 49.5 11 12.9% 11 12.9%

100–199 99.5–199.5 149.5 26 30.6% 37 43.5%

200–299 199.5–299.5 249.5 34 40.0% 71 83.5%

300–399 299.5–399.5 349.5 10 11.8% 81 95.3%

400–499 399.5–499.5 449.5 4 4.7% 85 100.0%

a. Calculate the real class limits and midpoint for each interval.

b. Construct a cumulative frequency, percentage, and cumulative percentage distribution.

c. What percentage of the cities have a property crime rate less than 300?

Cumulative % = 83.5%

14. The data below represent the crime rate per 10,000 for 25 cities and towns in the state of Maryland.

135 180 190 137 154 149 164 185 173 163 162 157 161

173 180 179 197 159 182 164 164 144 152 150 163

With this data construct a grouped frequency distribution. Use seven intervals, begin your first interval with a lower stated limit of 130. Also, show the real class limits, midpoints, frequencies, cumulative frequencies, proportions, cumulative proportions, percentages, and cumulative percentages. (Hint: The midpoint is simply the (lower limit + upper limit) / 2). Include your table and use it to answer the following questions:

Real Limits |Stated Limits |f |Mid |cf |Prop. |Cum

Prop. |% |Cum

% | |129.5–139.5 |130–139 |2 |134.5 |2 |0.080 |0.080 |8.0 |8.0 | |139.5–149.5 |140–149 |2 |144.5 |4 |0.080 |0.160 |8.0 |16.0 | |149.5–159.5 |150–159 |5 |154.5 |9 |0.200 |0.360 |20.0 |36.0 | |159.5–169.5 |160–169 |7 |164.5 |16 |0.280 |0.640 |28.0 |64.0 | |169.5–179.5 |170–179 |3 |174.5 |19 |0.120 |0.760 |12.0 |76.0 | |179.5–189.9 |180–189 |4 |184.5 |23 |0.160 |0.920 |16.0 |92.0 | |189.5–199.5 |190–199 |2 |194.5 |25 |0.080 |1.000 |8.0 |100.0 | |N | |25 | | | | | | | |

a. What is the width of your stated class limits?

10

b. How many cities had crime rates equal to or greater than 150 crimes per 10,000?

25 - 4 = 21

c. What percentage of cities had crime rates under 180 crimes per 10,000?

Cumulative% = 76%

d. What proportion of cities had a crime rate between 160 and 169 crimes per 10,000 people?

7 / 25 = .28

e. Describe in your own words why it is useful to create real class limits from stated class limits.

Real class limits transform ordinal data back into interval level data.

f.

[pic]

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download