Instructor:R. B. Taylor

Voice: 215.204.7169 Fax: 610.446.9023

E-mail: V1008E@VM.TEMPLE.EDU

Time:Wed. 6 p.m. - 8:30 p.m.

Where: Gladfelter 5^th floor classroom and lab

Office:539 Gladfelter

Office hours:Wed. 4 - 6 and by appointment

 

I will give you my home phone number. Please feel free to contact me about any problems, difficulties or questions you may have that relate to this class at any reasonable hour of the day or evening. This will be an extremely fast-paced course, so it is important that you contact me as speedily as possible should any concerns or difficulties arise.

 



This course concentrates on two multivariate techniques: hierarchical linear models and time series. The former are appropriate when units of observation are nested within broader units. The latter is appropriate when units are assessed repeatedly over time. You will learn the conceptual background for each technique, the special terms associated with each, how each applies to a range of criminal justice research and evaluation questions, and, perhaps most importantly, how to carry out these analyses and interpret the resulting output. You will complete weekly readings and written assignments. Almost all written assignments will require completing a computer lab assignment. There will be an in-class final, written examination.

 



Hierarchical Linear Models (HLM) represent a significant advance in social scientists' ability to understand how outcomes are affected by context, how individual and contextual factors interact, how outcomes change over time, and how to summarize results from a series of studies. We will be concentrating in this course largely on the first two uses of HLM. We will get to the other two uses of HLM - to investigate changes over time, and to summarize studies -- if and as time permits.

 

Hierarchical linear models address a range of theoretical and methodological issues relevant to criminal justice, sociology, psychology, urban studies, education, and political science. The issues include multilevel analysis, aggregation issues, contextual analysis, and clustered samples. In simple, whenever the individual units of study (e.g., students) are nested within a higher level unit (e.g., schools), HLM is an appropriate, and some would argue the most appropriate form of analysis.

 

Here are some examples of "units nested within larger units" in criminal justice evaluation or research:

 

Level 1 units	Level 2 units
Residents	Different Neighborhoods
Police Officers	Different Precincts
Police Precincts	Different Police Departments
Cases Sentenced	Different Judges
Prisoners	Different Prisons
Sentenced Drug Offenders	Different Drug Courts
Juveniles	Different Treatment Programs

In this class we use numerous examples from criminal justice, but also include a substantial number of examples from other disciplines. In the half of this course we will:

• Learn the conceptual and methodological issues addressed by HLM

   
  
  

• Examine the conceptual framework used by HLM

   
  
  

• Set up and run HLM analyses

   
  
  

• Interpret basic output generated by HLM

   
  
  The second tool we will examine this semester is time series analysis. Time series helps criminal justice researchers, evaluators, and policy makers understand outcome measures that may shift over time. These may include crime rates, prison admission rates, court caseloads, victimization rates, and more. For outcomes such as these time series helps us describe the changes over time in a mathematical fashion (modeling), forecast future scores on the outcome (prediction), and learn how much the trend has shifted as a result of a specific intervention (evaluation).

   
  
  For time series you will learn:

   
  
  

• How to describe a data series over time using regression techniques

   
  
  

• How to address the special problems for regression presented by time series data

   
  
  

• How to model a data series, and an impact evaluation

   
  
  

• How to gauge the "quality" of your model

   
  
  The goals of this course revolve around three different sets of activities. First, we hope to learn the rudimentary conceptual frameworks underlying these analyses; what problems does the analysis identify, and how does it propose to address these concerns? Second, we want to see how these tools have been applied to actual research problems, in criminal justice and in related fields. Third, we want to have "hands on" experience with the techniques themselves, using available datasets.

   
  
  We will be spending some time for each class in the classroom. In addition, in almost every class, we will be spending some time in the 5^th floor computer lab.

   
  
  We will be using two sets of statistical programs. For HLM we will use a specific program, put out by Scientific Software International. The department has bought a site license for this program. For the time series analysis we will use the series module that is part of SPSS. In addition, to examine and learn about variables in datasets, we will rely upon SPSS base module capabilities.

   
  
  In terms of datasets, we will be focusing attention for the weekly assignments on several datasets; I will probably put these on one or more machines in the lab. One of the data sets comes with HLM and looks at math achievement of different students in different kinds of schools. I strongly suggest that you copy all of the data sets to floppies and make those floppies read only, so you can be sure you have clean copies of the data files.

   
  
  I assume you understand the basics of OLS multiple regression, including:

   
  
  variance

  correlation

  scatterplots

  R squared

  adjusted R squared

  F test of R squared

  b weights

  standard errors of b weights

  beta weights

  t tests of b weights

  constant

  residuals

  predicted scores

  residual diagnostics

  tests for linear vs. curvilinear impacts

   
  
  In addition, I assume you know your way around SPSS for Windows, either version 6.0, 6.1, 7.5 or 8.

   
  
  

  You will be reading books, articles, and handouts.

   
  
  You should find the following books in the Temple SWEATSHIRT and bookstore:

   
  
  Bryk, A., and Raudenbush, S. (1992) Hierarchical linear models. Thousand Oaks: Sage

   
  
  McDowall, D., McCleary, R., Meidinger, E. E. and Hay, Ra. A., Jr. (1980). Interrupted time series analyses (Quantitative Applications in the Social Sciences, No. 21). Beverly Hills: Sage.

   
  
  Ostrom, C. W. Jr. (1978). Time series analysis: Regression techniques. (Quantitative Applications in the Social Sciences, No. 9). Beverly Hills: Sage.

   
  
  SPSS, Inc. (1994). SPSS Trends 6.1. Chicago: SPSS.

   
  
  For the readings, I have not decided yet if I will try and put these together into a packet, that you can buy, or if we will make another arrangement.

   
  
  I have an extensive series of handouts on HLM that I will get to you.

   
  
  
  
  You will complete the assigned readings and problems on a weekly basis and come to class prepared.

   
  
  You will complete and write up each weekly analytical assignment. These weekly papers comprise a substantial portion of your grade. The assignment will be handed out on WEDNESDAY. Your completed papers are due in my office or mailbox absolutely no later than 2 p.m. the following MONDAY. That way I will have time to review each completed assignment, and grade it, and get it back to you so that we can discuss it in the next class period.

   
  
  These papers are to be

• typed, double-spaced

• with only the due date and your SSN at the top; I do not want to see your name anywhere

   
  
  Be sure to spell-check and grammar-check your papers, since I do take off for flagrantly poor grammar, and for mis-spelled words.

   
  
  The readings provide the needed conceptual background for carrying out the work assigned. Thus it is important that you keep up with the readings.

   
  
  You will notify me beforehand if it is absolutely essential for you to miss a class. Given the amount of ground we must cover, a missed class may create a significant burden for your learning curve. If you do miss a class it is completely your responsibility to get all handouts, assignments, and so on, that were distributed.

   
  
  Every now and then I will hand out some questions about the readings assigned for that week. I expect you, by the next class, to have attempted to answer some of those questions. Your answers should be typed, double-spaced, with both your name and SSN on the top of the page. To get credit for answering these questions, your answers should be turned in by the beginning of the following class. I am not grading your answers, but rather just giving you credit for trying to answer them. I am not sure for how many weeks we will do these.

   
  
  Your grade at the end of the semester will be based on the following:

   
  
  70%Average grade on all weekly written home works

   
  
  25%Final examination. This will focus on the identification of an appropriate tool to use in a particular situation; and on interpreting results presented

   
  
  5%Turned in answers to questions on readings. These questions will be handed out intermittently through the semester.

   
  
  
  
  1. Assignments are due on the date indicated. I reserve the right to lower the grade for assignments that are handed in late. The amount the grade is lowered will increase the longer the delay in handing the assignment in.

   
  
  2. If you have an excuse for a late assignment I will take this in to account only if you notify me -- or try to notify me -- beforehand about the problem and I find your excuse for the delay to be a valid one (e.g., car accident).

   
  
  3. We will discuss in class the nature of academic misconduct, including plagiarism. You are responsible for understanding the different varieties of academic misconduct. If I encounter solid evidence of academic misconduct I reserve the right to fail you on the assignment in question, and/or to assign you a failing grade for the course. I will try to state as clearly as I can the ways in which it is acceptable for you to cooperate with one another and network, and the ways in which it is not acceptable.

   
  
  4. You do have a right to submit any assignment for regrading. You should state in writing the reason you think you deserve a higher grade, attach that to the original completed assignment, and return it to me. Your grade may go up, go down, or stay the same. I may consult with other faculty members as I deem fit.

   
  
  5. Of course, on any grading issue that you and I are unable to resolve to our mutual satisfaction you do have available to you standard grievance procedures you may elect to pursue. You should contact the chair of the graduate program for more details.

   
  
  
  
  The sequence of topics and readings, as best as I can predict them at this point, appear below. All of this is subject to change depending upon numerous factors, including el Nino.

   
  
  

  Class Date	Topics / Readings handed out / Other
  1/21	Syllabus review; Introduction to data sets; Assumptions of OLS violated with clustered data; What is a contextual regression; What are the problems accompanying contextual regression; The aggregation problem Readings: Thorndike 1939; Covington & Taylor 1991; Hauser 1970; Hauser 1974; Sampson 1993 HLM manual pages HLM Handout: HLML01 Assignment 1: Conduct and interpret a contextual regression
  1/28	Interpreting contextual regressions Introduction to HLM: The Level 1 model; Variances and Covariances; Variance Decomposition; Empirical Bayes estimates; "true" scores on the group means; HLM SUBMODEL 1: One-way ANOVA with random effects; HLM SUBMODEL 2: One-way ANCOVA with random effects Readings: B&R 1-19,60-64 ; Kurtz & Taylor 1997; Liska 1990; Arnold 1992 HLM Handout: HLML02 Assignment 2: Conduct and interpret a variance decomposition; complete and interpret a Level 1 model (One-way ANCOVA with random effects)
  2/4	HLM SUBMODEL 3: Random coefficients regression model Interpreting the Level 2 portion of the model; HLM SUBMODEL 4: Means as outcomes regression Centering at Level 1 & 2; Intraclass correlation Readings: B&R 20-31; 64-70; Shrout & Fleiss 1979 HLM Handout: HLML03, HLML04, HLML05 Assignment 3: complete and interpret means as outcomes regression
  2/11	HLM FULL MODEL: Intercepts and Slopes as Outcomes (IASAO) Residuals: Doing Diagnostics; Variables in the file; Plots you want to generate; Readings: Bryk & Thum 1989; Rountree & Land 1996; Sampson, Raudenbush & Earls 1997; Taylor 1997 B&R: 70-82; 197-224 HLM Handout: HLML10, HLM11 Assignment 4: complete and interpret full model with fixed slopes
  2/18	More on residual diagnostics; general catch-up session Assignment 5: report on residuals analysis
  2/25	Introducing variation over time; HLM and criminal careers; HLM and trends Readings: Raudenbush & Chan 1992; Osgood & Smith 1995 B&R: Chapter 6 HLM Handout: HLML06, HLML07 Assignment 6: HLM analysis of crime changes over time
  3/4	More on HLM and changes over time
  3/11	NO CLASS - BREAK
  3/18	Introduction to time series data Descriptive information about time series; plots and interpretation Exponential smoothing - Introduction Readings: SPSS, Inc. (1994), Chapter 1 - 4 Flango & Elsner 1983; Gardner 1985 Assignment 7: Report on descriptives and plots
  3/25	Exponential smoothing - More Assignment 8: Exponential smoothing
  4/1	Introduction to curvefit (regression) models Readings: SPSS, Inc. (1994), chapters 5 - 9 Ostrom (1978) Blumstein & Moitra 1979; Landau & Pfefferman 1988 Assignment 9: Curvefit models
  4/8	Univariate ARIMA models: Introduction Readings: McDowall et al. 9 - 64 SPSS, Inc., chapter 6 Cantor & Land 1991; Carr-Hill 1992; Farrell & Pease 1994; Hale 1989; Hale & Sabbagh 1991a; Hale & Sabbagh 1991b;
  4/15	Univariate ARIMA models: Continued Readings: SPSS, Inc., chapters 7-8 Myers 1995; McCain & McCleary 1979 Assignment 10: Univariate ARIMA analysis: Modeling Philadelphia Arrests
  4/22	ARIMA models: Impact assessment Readings: McDowall et al. (1980) 65 - 93 Berk, Loseke, Berk et al. 1982; Berk, Rauma, Loseke et al. 1984; Deutsch 1979; Deutsch & Alt 1977; Jones 1990; Kaminski, Edwards & Johnson 1997; Loftin, Heumann & McDowall 1983 Assignment 11: Impact assessment
  4/29	Diagnostics, residuals, issues of fit
  5/6	Final Examination - in class

   
  
  
  

  References

   
  
  Arnold CL (1992) An Introduction to hierarchical linear models. Measurement and Evaluation in Counseling and Development 25:58-90

   
  
  Berk RA, Loseke DR, Berk SF, Rauma D (1982) Bringing the cops back in: A Study of efforts to make the criminal justice system more responsive to incidents of family violence. Social Science Research 9:193-215

   
  
  Berk RA, Rauma D, Loseke LR, Berk SF (1984(?)) Throwing the cops back out: The Decline of a local program to make the criminal justice system more responsive to incidents of domestic assault. Social Science Research 11:245-279

   
  
  Blumstein A, Moitra S (1979) An Analysis of the time series of the imprisonment rate in the United States: A Further test of the stability of punishment hypothesis. Journal of Criminal Law and Criminology 70:376-390

   
  
  Bryk AS, Thum YM (1989) The Effects of high school on dropping out: An exploratory investigation. American Educational Research Journal 26:353-384

   
  
  Cantor D, Land K (1991) Exploring possible temporal relationships of unemployment and crime: A Comment on Hale and Sabbagh. Journal of Research in Crime and Delinquency 28:418-425

   
  
  Carr-Hill RA (1992) Review of S. Field "Trends in crime and their interpretation". British Journal of Criminology 32:222-225

   
  
  Covington J, Taylor RB (1991) Fear of crime in urban residential neighborhoods: Implications of between and within-neighborhood sources for current models. Sociological Quarterly 32:231-249

   
  
  Deutsch SJ (1979) Lies, damn lies, and statistics. Evaluation Quarterly 3:315-328

   
  
  Deutsch SJ, Alt FB (1977) The effect of Massachusetts' gun control law on gun-related crimes in the city of Boston. Evaluation Quarterly 1:543-568

   
  
  Farrell G, Pease K (1994) Crime seasonality: Domestic disputes and residential burglary in Merseyside, 1989-90. British Journal of Criminology 34:487-498

   
  
  Flango V, Elsner ME (1983) Estimating caseloads: Two methods tested in Tulsa. State Court Journal Spring:18-24

   
  
  Gardner ES (1985) Exponential smoothing: The State of the art. Journal of Forecasting 4:1-28

   
  
  Hale C (1989) Unemployment, imprisonment, and the stability of punishment hypothesis: Some results using cointegration and error correction models. Journal of Quantitative Criminology 5:169-186

   
  
  Hale C, Sabbagh D (1991a) Testing the relationship between umemloyment and crime: A Methodological comment and empirical analysis using time series data for England and Wales. Journal of Research in Crime and Delinquency 28:400-417

   
  
  Hale C, Sabbagh D (1991b) Unemployment and crime: Differencing is no substitute for modeling. Journal of Research in Crime and Delinquency 28:426-429

   
  
  Hauser PM (1970) Context and consex. American Journal of Sociology:645-664

   
  
  Hauser PM (1974) Contextual analysis revisited. SMR 2:365-375

   
  
  Jones PR (1990) Community corrections in Kansas: Extending community-based corrections or widening the net? Journal of Research in Crime and Delinquency 27:79-101

   
  
  Kaminski RJ, Edwards SJ, Johnson JW (1997) The Deterrent effects of Oleoresin capsicum on assaults against police.

   
  
  Kurtz E, Taylor RB (1997) Outcome variance decomposition of responses to crime with hierarchical linear models.

   
  
  Landau SF, Pfefferman D (1988) A Time series analysis of violent crime and its relation to prolonged states of warfare: The Israeli case. Criminology 26:489-504

   
  
  Liska AE (1990) The Significance of aggregate dependent variables and contextual independent variables for linking macro and micro theories. Social Psychology Quarterly 53:292-301

   
  
  Loftin C, Heumann M, McDowall D (1983) Mandatory sentencing and firearms violence: Evaluating an alternative to gun control. Law and Society Review 17:288-317

   
  
  McCain LJ, McCleary R (1979) The Statistical analysis of the simple interrupted time series quasi-experiment. In: Cook T, Campbell D (eds) Quasi-Experimentation. Rand McNally, Chicago

   
  
  Myers M (1995) Gender and southern punishment after the Civil War. Criminology 33:17-46

   
  
  Osgood W, Smith G (1995) Applying hierarchical linear modeling to extended longitudinal evaluations: The Boys Town follow-up study. Evaluation Review 19:3-28

   
  
  Raudenbush SW, Chan W (1992) Growth Curve Analysis in Accelerated Longitudinal Designs. Journal of Research in Crime and Delinquency 29:387-411

   
  
  Rountree PW, Land KC (1996) Burglary victimization, perceptions of crime risk, and routine activities: A Multilevel analysis across Seattle neighborhoods and census tracts. Journal of Research in Crime and Delinquency 33:147-180

   
  
  Sampson RJ (1993) Linking Time and Place: Dynamic contextualism and the Future of Criminological Inquiry. Journal of Research in Crime and Delinquency 30:426-444

   
  
  Sampson RJ, Raudenbush SW, Earls F (1997) Neighborhoods and violent crime: A Multi-level study of collective efficacy. Sci

   
  
  Shrout PE, Fleiss JL (1979) Intraclass correlations: Uses in assessing rater reliability. Psychological Bulletin 86:420-428

   
  
  Taylor RB (1997) Social order and disorder of streetblocks and neighborhoods: Ecology, microecology and the systemic model of social disorganization. Journal of Research in Crime and Delinquency 33:113-155

   
  
  Thorndike GL (1939) On the fallacy of imputing the correlations found for groups to the individuals in smaller groups composing them. American Journal of Psychology 52:122-124