Spring 2014

EPsy 8262

Regression and the General Linear Model

EPsy 8262 is the second course of the two-semester Ph.D. level statistics sequence in Educational Psychology at the University of Minnesota. The course will cover a number of regression methods. Emphasis will be placed on viewing traditional statistical methods as special cases of multiple regression, which itself is a special case of the general linear model (GLM). Emphasis in the course is given to learning how to do data analysis. We will also devote considerable time to illustrating how to present regression results in prose, tables and figures. Instead of formal textbook readings, each unit is supported by lectures and course notes.

Instructor Information

Instructor: Andrew Zieffler

Office: Education Sciences Building 163 [map]

Office Hours: Tuesday, 9:00a.m–10:00a.m.; and by appointment

Email: zief0002@umn.edu

Teaching Assistant

The teaching assistant for EPsy 8262 this semester is Laura Ziegler.

Office: Education Sciences Building 192 [map]

Office Hours: Monday, 11:00a.m–12:00p.m.; Wednesday, 11:45a.m–12:45p.m.; and by appointment

Email: sath0166@umn.edu

Audience and Course Prerequisites

It is assumed that the student has facility with algebra. It is also assumed that the student has taken EPsy 8261 or an equivalent course. Lastly, it is assumed that students enrolled in EPsy 8262 have familiarity with using computers and technology (e.g., internet browsing, Microsoft Word, opening/saving files, etc.).

If your knowledge of basic statistics is rusty, please review the principles of estimation and inference in an introductory statistics textbook. If you have not successfully completed an introductory statistics course, EPsy 8262 is not the right course for you.

Please inquire if you have any questions about whether your statistics background is sufficient.

Course Philosophy

In my mind, statistics is more than a particular methodology used in different disciplines; it is more than a just an application of mathematics. Statistics is a principled way of thinking about the world, in particular it is a principled approach to data collection, prediction, and scientific inference.

Statistics is a discipline that has, like many others, undergone a tremendous amount of growth and change in the last two decades. In today's dynamic and interdisciplinary world, success in confronting new analytical issues requires both substantial knowledge of a scientific or technological area and highly flexible problem-solving strategies.

Internalizing a disciplines' way of thinking about and solving problems is a time consuming process, with the keyword being “process”. It is not something that can be taught to students in a semester or even year-long course. Learning statistics takes much more than memorizing formulae or R functions. It requires active participation and thought in and out of the classroom. I can pose questions, direct you to resources, provide you with opportunities to learn the material, and impart what little wisdom I have, but in the end, you will have to do all of the hard work.

Course Content

EPsy 8262 is structured around 11 learning units (shown below). Some will take just one class session; others will take two or three. Specific learning objectives, notes, R scripts, and course readings are provided for each unit.

Introduction to Statistical Modeling and Regression

In this set of notes you will learn about: describing research in terms of research questions, variables, and predictors; differences between statistical models and deterministic models; how to begin answering research questions about relationships between variables from the examination of predictor and outcome distributions and scatterplots; how to mathematically represent the population model and interpret its components; how to fit a hypothesized model to data using the method of least squares; model–data residuals; uses of the fitted regression model for helping to improve predictions; explained variation; and using the analysis of variance to estimate the mean square error (MSE).

Please read the following blog post.

Here are the notes and R script for the unit.

For those students interested in learning more about ggplot2, here are some references for you.

Correlation and Causality

In this set of notes you will learn about the correlation coefficient (r) by first developing a heuristic understanding of it, then by seeing correlation as regression on standardized variables, and finally through understanding the relationship between correlation and explained variation. We then learn about necessary conditions required for moving beyond correlation to causation including why randomized experiments are the “gold standard” for establishing causality; what can be done when randomized experiments are not possible; when an observed correlation might not indicate a causal relationship (spurious correlation, confounding, Simpson's paradox, reciprocal causation and ecological correlation).

Please read the article by Goodwin, & Leech (2006).

Here are the notes and R script for the unit.

Inference for the Regression Model

In this set of notes you will learn to distinguish between population models and fitted sample results. We will address questions such as: How would regression results differ upon repeated random sampling from the population? What would the sampling distribution of a regression parameter look like? How can we estimate the sampling distribution with results from just a single sample? You will learn how to test statistical hypotheses about regression parameters and also how to find confidence intervals for these parameters. You will also learn how to find a confidence interval for the mean of Y at a given value of X. Lastly, you will learn how to find a prediction interval for an individual value of Y at a given value of X.

Please read the article by Gopen & Swan (1990).

Here are the notes and R script for the unit.

Here are a couple of readings to get you in high Valentine's Day spirits this week. The first article, published in Social Science & Medicine, examined the effects of cultural representations, in the form of salient holidays, on birth timing. In the second article, published in Advances in Consumer Research, the authors expand upon an earlier finding (presented in a conference paper) that men have different attitudes toward Valentine's Day and Valentine's Day gift-giving than women. The qualitatively examine men's beliefs about the purpose of Valentine's Day, what they like most and least about the holiday, and why they did or did not participate in gift-giving activities. For those of you that are a little more cynical about the holiday, perhaps you might enjoy the third article describing the 1929 massacre of Bugs Moran and his crew that was published in the American Journal of Forensic Medicine & Pathology.

Assumptions for the Regression Model

In this set of notes you will learn about the assumptions required for least squares estimation and inference. We will cover four major types of model violations: (1) outliers; (2) non-linearity; (3) heteroscedasticity; and (4) non-independence of errors. You will learn that residuals are controlled observations and examine why residuals provide a powerful lens for evaluating regression assumptions. You will also learn about both raw and studentized residuals. You will learn how to construct residual plots and how to evaluate these plots for potential problems. Lastly, you will gain some insight in to how to deal with potential outliers or other unusual observations.

Here are the notes and R script for the unit.

For students that would like additional practice, and for interested foodies, here are the data from the 2008 and 2011 Zagat rated restaurants.

Transformations to Achieve Linearity

In this set of notes you will learn about what happens when a linear regression model is fitted to data that are nonlinearly related. You will also learn abut three alternative statistical models that are useful for nonlinear relationships. You will be given a brief refresher on the mathematics of logarithms, and the effects of logarithmic transformations on variables. You will then learn how to use logarithmic transformations to model nonlinear relationships, including the interpretations of the parameters from these models. You will also learn about the differences between taking logarithms to base 2, 10, and e. Lastly, you will learn the “the Rule of the Bulge” for selecting among alternative transformations.

If you would like a refresher on logarithms, check out the link below. For more on logarithmic transformations, especially in multiple regression, you can also read Example 6 from Chapter 3 of Edward Tufte's book Data Analysis for Politics and Policy.

Here are the notes and R script for the unit.

The Basics of Multiple Regression

In this set of notes you will learn about various representations of the multiple regression model including: (1) an algebraic model; (2) a graphical representation in three-dimensions; and (3) a graphical representation in two-dimensions. You will also learn the basics of how multiple regression works, how estimates of the parameters are obtained, and what it means to hold predictors “constant”. You will also learn how to interpret and plot the results of a multiple regression analysis. Lastly, you will learn about inference for both the model and individual parameters in multiple regression analyses.

Please read the article by Charles, Dinwiddle, & Massey (2004).

Here are the notes and R script for the unit.

Statistical Control: Correlation and Collinearity

In this set of notes, yuo will learn: What is really meant by statistical control and whether statistical control is always possible. You will learn how to scrutinize a correlation matrix for what it may foreshadow for multiple regression. We will use Venn diagrams to develop your intuition about correlation and help you further understand partial correlation, including terminology, interpretation, and relationship to simple correlation. We will also take a more extensive look at multiple correlation. We will also talk about suppressor effects and when statistical control can help reveal an effect. We will examine the problem of collinearity: What it is; How to spot it; and What to do about it.

Please read the article by Berkman & Plutzer (2004).

Here are the notes and R script for the unit.

Categorical Predictors I: Dichotomies

In this set of notes you will learn about the unusual marriage between regression and categorical predictors and regression. You will learn how to create and name dummy (or indicator) variables. You will also learn how to include a dummy variable in a multiple regression model; how and why it operates; what happens if we change the reference category; and how regressing Y on a dummy variable directly relates to the two-sample t-test. You will also continue to use both the simple and partial correlation matrices to foreshadow regression results. Lastly, we will again examine how to present results from these models including presentation of the adjusted means, graphic displays of regression findings, and the display and interpretation of prototypical trajectories.

Please read the article by Wendorf (2004).

Here are the notes and R script for the unit.

Categorical Predictors II: Polychotomies

In this set of notes you will learn about distinguishing between nominal and ordinal predictors. You will also learn how a series of dummy variables can represent a polychotomous nominal predictor. We will lokk at why regressing Y on all but one dummy variable yields the desired model, and the consequences of changing the reference category both for parameter estimates and hypothesis tests. We will also revisit the problem of multiple comparison and look at one solution to this problem, namely the Bonferroni multiple comparison procedure. We will also examine an alternative way of getting the identical results, the analysis of variance (ANOVA). We will also look at a model for using an ordinal predictor. We again look at how to present results from these models including the presentation of adjusted means and graphs. Finally, we will untangle the nomenclature of regression, analysis of variance, and analysis of covariance.

Here are the notes and R script for the unit.

Interaction and Quadratic Effects

In this set of notes we will answer the question: what is a statistical interaction and learn how it is different from the effects we have modeled so far. We will briefly talk about differences between ordinal and disordinal statistical interactions, and we will learn how to test for the presence of a statistical interaction. In learning about interaction effects, we will mathematically examine why including a cross-product term will tell us about a statistical interaction. We will also look at two different ways of summarizing interaction effects and how the interaction model compares to fitting separate models within groups. We will also study a very special interaction—the interaction of a predictor with itself (quadratic regression model). We will also look at the relationship between the quadratic model and the logarithmic models we fit earlier in the course.

Please read the article by Baron & Kenny (1986).

Here are the notes and R script for the unit.

Regression Modeling in Practice

In this unit we will put into practice many of the concepts we have covered in the course.

Please complete the preparation assignment prior to the in-class portion of this unit.

Course Textbook and Readings

One of the following textbooks is highly recommended for the course.

I also recommend the following books as desk references for APA writing and publication. Aside from the Publication Manual of the American Psychological Association (Sixth Edition)—which is a necessity—the three companion books provide additional guidance on writing and creating figures and tables for APA.

For students wanting reference books for learning R, Teetor (2011) is a great option, as is Chang (2012). The former is more general, while the latter is an indispensible resource for learning how to create plots using ggplot.

Lastly, there are several articles that you will need to read during the semester (TBA on the Calendar section of the course website). These can be obtained through the Journals link on the University of Minnesota Libraries website. In order to access the full text of the articles, you may need to log in using your University x500 username and password.

Course Requirements and Student Evaluation


Homework will make up 100% of your course grade. There are 10 homework assignments, each making up an equal proportion of the grade. The homework assignments will be posted on the course website.

Students will be expected to develop proficiency in writing coherent summaries and interpretations of data analyzed by the methods introduced in the course. The homework assignments include problems that will help students learn the course material and software through reflection and practice.

Submitted homework assignments must be typed—handwritten assignments will ordinarily receive no credit. Homework assignments that are submitted via e-mail without prior instructor approval will receive no credit. If approval is granted to turn in an assignment via e-mail the only acceptable format is a PDF file.

To foster cooperation and collaboration, you are permitted to form groups of no larger than three to work on the homework assignments. For all work handed in, list the names of the group members in alphabetical order. Each assignment will be assigned a grade and this grade will be applied to the individuals within the group.


Students are expected to actively participate in the course. Active participation includes, but is not limited to, being engaged during the class, asking questions, providing additional insight and material, responding to other students and the instructor, and always being open and inquisitive. While not explicitly a part of the course grade, your participation in the course will play a role if you are between grades at the end of the semester.

Student Evaluation

Course grades will be based entirely on the homework assignments. If you are taking the course S/N, the minimum criterion to receive an S is 80% (the equivalent of a B– letter grade).

  • A (94%)
  • A– (90%)
  • B+ (87%)
  • B (84%)
  • B– (80%)
  • C+ (77%)
  • C (74%)
  • C– (70%)
  • D (60%)
  • F (Below 60%)

Shortly after the course, you may access your grades on-line at http://www.onestop.edu. Homework assignments will be handed back in class or during office hours. Uncollected homework assignments and course projects will be retained for three weeks after the course and then discarded.

An incomplete for this course will be considered on a case-by-case basis. The most valid reason for an incomplete is an unforeseen event that gravely interferes with a student's ability to perform at an adequate level. An incomplete will not be given for unqualified poor performance.

Student Collaboration

You do not need to join a homework group to be successful in this course, however most students find it helpful. Please choose your work group partners carefully as I am not willing to manage intragroup conflicts or assign varying grades within a group. If you are taking the course as the S/N grading option, I strongly discourage you from joining a group unless the others members are also taking the course as S/N. If you are auditing the course I forbid you from joining a group unless your group consists of other auditors (auditors hand in no work).

How Can I Be Successful in this Course?

There are several things you can do to be successful in this course. First and foremost, complete all of the readings and come prepared to class. Work with a group on the homework assignments, but attempt the assignments alone before meeting with your group.

If you are experiencing problems, need help, or have any questions or other course-related concerns, do not hesitate to get in touch with the instructor.

Statistical Computing

Statistical computing is an integral part of statistical work and subsequently EPsy 8220. To support your learning in this area, this course will emphasize the use of R. R is a free software environment for statistical computing and graphics. It compiles and runs on a wide variety of UNIX platforms, Windows and MacOS (http://www.r-project.org). It is assumed that everyone enrolled in the course is comfortable using a computer to perform basic statistical analysis (although it is not assumed that you have used R).

While some R syntax and programming is taught during class time, there is also a fair amount that you may need to learn on your own outside of class. There are several tutorials and resources available on the web to help you learn R at your own pace.

Downloading and Installing R

In order to download and install R your computer must be connected to the Internet. The latest version of R can be obtained from the R Project for Statistical Computing.

After navigating to the website click on “CRAN” under “Download, Packages” on the left-hand side of the welcome screen. You must choose a server in your country of origin, called a CRAN mirror. After doing so, select the appropriate operating system for your computer–Linux, MacOS, or Windows. For Linux and MacOS, follow the directions at the top of the download page. For Windows, download the base package and install it like any other executable file. (On Windows machines you might need to have “administrator” privileges to successfully install and use the program.)

Computing Resources

Here are some free online resources for learning R:

  • Comparing Groups (Chapters 1 & 2): These chapters will introduce you to basic R syntax [Chapter 1] and data manipulation [Chapter 2]
  • Using R for Data Analysis: This book chapter by Ken Kelley, Keke Lai, and Po-Ju Wu gives a gentle introduction to R for researchers in the social sciences [online access].
  • SimpleR: This 114 page reference will illustrate how to use R to carry out most analyses covered in an introductory statistics course [online access].
  • Contributed Documentation: These are manuals, tutorials, etc. provided by users of R. Access this on the R website by clicking the “Other” link under “Documentation” and then clicking “Contributed Documentation” [online access].


RStudio is an integrated development environment (IDE) for R. RStudio combines an intuitive user interface with powerful coding tools to help you get the most out of R. RStudio Desktop is free and can be downloaded at http://www.rstudio.org/download/.

Course Technology Policy

The course uses technology on a regular basis during both instruction and assessments (e.g., homework assignments, exams, etc.). Student difficulty with obtaining or operating the various software programs and technologies—including printer trouble—will not be acceptable as an excuse for late work. Due to the variation in computer systems and the difficulty in assessing problems via email, the instructor and/or TA may not be able to assist in trouble shooting all problems you may have. In these cases contact the university Academic and Distributed Computing Services (ADCS) or your systems administrator (if you have one).


Email is the primary source of communication among instructors, teaching assistants, and students for this course. As such, you will be expected to check your email frequently (i.e., at least once per day). As per the University of Minnesota policy,

Students are responsible for all information sent to them via their University assigned email account. If a student chooses to forward their University email account, he or she is responsible for all information, including attachments, sent to any other email account.

Course Website

Most of the homework assignments, data files, etc. are available on the course website (http://www.tc.umn.edu/zief0002/8220.php). The website works best with a recent version of Mozilla Firefox, Google Chrome, or Safari.

Mac Users

If you are using a Mac and seem to have problems downloading files, hold the option-key while clicking on the link. This should download the file to your desktop. You then need to erase the .txt suffix that is appended to the end of the file. For example, a comma separated value (CSV) file should have the suffix .csv, and not .csv.txt.

Use of Personal Electronic Devices in the Classroom

Using personal electronic devices in the classroom setting can hinder instruction and learning, not only for the student using the device but also for other students in the class. To this end, the University establishes the right of each faculty member to determine if and how personal electronic devices are allowed to be used in the classroom. See http://policy.umn.edu/Policies/Education/Education/CLASSROOMPED.html.

Campus Computer Labs

The Office of Information Technology (OIT) manages numerous computer labs on the Twin Cities campus. Students from all colleges may drop in to use the computer labs during open hours. The OIT website contains information pertaining to the location, hours, and software available for each of the computer labs.

Mission Statements

Quantitative Methods in Education

The Quantitative Methods in Education (QME) track offers educational opportunities in both quantitative and qualitative methods with a broad array of introductory and advanced coursework. Students who choose QME as their track within educational psychology may specialize in any of four areas: measurement, evaluation, statistics, and statistics education. The goal of QME is to provide students with broad but rigorous methodological skills so that they may conduct research on methodologies, may help to train others in methodology, or will have the skills necessary to conduct research in related fields.

Department of Educational Psychology

Educational psychology involves the study of cognitive, emotional, and social learning processes that underlie education and human development across the lifespan. Research in educational psychology advances scientific knowledge of those processes and their application in diverse educational and community settings. The department provides training in the psychological foundations of education, research methods, and the practice and science of counseling psychology, school psychology, and special education. Faculty and students provide leadership and consultation to the state, the nation, and the international community in each area of educational psychology. The department's scholarship and teaching enhance professional practice in schools and universities, community mental health agencies, business and industrial organizations, early childhood programs, and government agencies. Adopted by the Department. of Educational Psychology faculty October 27, 2004.

College of Education and Human Development

The new College of Education and Human Development is a world leader in discovering, creating, sharing, and applying principles and practices of multiculturalism and multidisciplinary scholarship to advance teaching and learning and to enhance the psychological, physical, and social development of children, youth, and adults across the lifespan in families, organizations, and communities.

University of Minnesota Policies and Procedures

Academic Freedom and Responsibility

Academic freedom is a cornerstone of the University. Within the scope and content of the course as defined by the instructor, it includes the freedom to discuss relevant matters in the classroom. Along with this freedom comes responsibility. Students are encouraged to develop the capacity for critical judgment and to engage in a sustained and independent search for truth. Students are free to take reasoned exception to the views offered in any course of study and to reserve judgment about matters of opinion, but they are responsible for learning the content of any course of study for which they are enrolled.* Reports of concerns about academic freedom are taken seriously, and there are individuals and offices available for help. Contact the instructor (Andrew Zieffler; zief0002@umn.edu), the Department Chair (Geoff Maruyama; geoff@umn.edu), your adviser, the associate dean of the college (Kenneth R. Bartlett; bartlett@umn.edu), or the Vice Provost for Faculty and Academic Affairs in the Office of the Provost (Arlene Carney; carne005@umn.edu).

*Language adapted from the American Association of University Professors “Joint Statement on Rights and Freedoms of Students”.

Disability Accommodations

The University is committed to providing quality education to all students regardless of ability. Determining appropriate disability accommodations is a collaborative process. You as a student must register with Disability Services and provide documentation of your disability. The course instructor must provide information regarding a course's content, methods, and essential components. The combination of this information will be used by Disability Services to determine appropriate accommodations for a particular student in a particular course. For more information, please reference Disability Services: http://ds.umn.edu

Equity, Diversity, Equal Opportunity, and Affirmative Action

The University will provide equal access to and opportunity in its programs and facilities, without regard to race, color, creed, religion, national origin, gender, age, marital status, disability, public assistance status, veteran status, sexual orientation, gender identity, or gender expression. For more information, please consult Board of Regents Policy: http://www1.umn.edu/regents/policies/ administrative/Equity_Diversity_EO_AA.html

Mental Health Services

As a student you may experience a range of issues that can cause barriers to learning, such as strained relationships, increased anxiety, alcohol/drug problems, feeling down, difficulty concentrating and/or lack of motivation. These mental health concerns or stressful events may lead to diminished academic performance and may reduce your ability to participate in daily activities. University of Minnesota services are available to assist you. You can learn more about the broad range of confidential mental health services available on campus via the Student Mental Health Website: http:// www.mentalhealth.umn.edu

Respecting Intellectual Property

Students may not distribute instructor-provided notes or other course materials, except to other members of the same class or with the express (written) consent of the instructor. Instructors have the right to impose additional restrictions on course materials in accordance with copyright and intellectual property law and policy. Students may not engage in the widespread distribution or sale of transcript-like notes or notes that are close to verbatim records of a lecture or presentation. For additional information, please see: http://policy.umn.edu/Policies/Education/Education/STUDENTRESP.html

Scholastic Dishonesty

You are expected to do your own academic work and cite sources as necessary. Failing to do so is scholastic dishonesty. Scholastic dishonesty means plagiarizing; cheating on assignments or examinations; engaging in unauthorized collaboration on academic work; taking, acquiring, or using test materials without faculty permission; submitting false or incomplete records of academic achievement; acting alone or in cooperation with another to falsify records or to obtain dishonestly grades, honors, awards, or professional endorsement; altering, forging, or misusing a University academic record; or fabricating or falsifying data, research procedures, or data analysis. (Student Conduct Code: http://regents.umn.edu/sites/default/files/policies/Student_Conduct_Code.pdf) If it is determined that a student has cheated, he or she may be given an “F” or an “N” for the course, and may face additional sanctions from the University. For additional information, please see: http://policy.umn.edu/Policies/Education/Education/INSTRUCTORRESP.html

The Office for Student Conduct and Academic Integrity has compiled a useful list of Frequently Asked Questions pertaining to scholastic dishonesty: http://www1.umn.edu/oscai/integrity/student/index.html. If you have additional questions, please clarify with your instructor for the course. Your instructor can respond to your specific questions regarding what would constitute scholastic dishonesty in the context of a particular class—e.g., whether collaboration on assignments is permitted, requirements and methods for citing sources, if electronic aids are permitted or prohibited during an exam.

Senate Academic Workload Policy

One conventional credit is hereby defined as equivalent to three hours of learning effort per week, averaged over an appropriate time interval, necessary for an average student taking that course to achieve an average grade in that course. It is expected that the academic work required of graduate and professional students will exceed three hours per credit per week or 45 hours per semester.

Senate Grading Policy

The University of Minnesota's grading policy is available online. For additional information, please refer to http://policy.umn.edu/Policies/Education/Education/GRADINGTRANSCRIPTS.html.

The University utilizes plus and minus grading on a 4.000 cumulative grade point scale in accordance with the following:

A 4.000 Represents achievement that is outstanding relative to the level necessary to meet course requirements
A– 3.667
B+ 3.333
B 3.000 Represents achievement that is significantly above the level necessary to meet course requirements
B– 2.667
C+ 2.333
C 2.000 Represents achievement that meets the course requirements in every respect
C– 1.667
D+ 1.333 Represents achievement that meets the course requirements in every respect
D 1.000 Represents achievement that is worthy of credit even though it fails to meet fully the course requirements
S Represents achievement that is satisfactory, which is equivalent to a C– or better
F/N Represents failure (or no credit) and signifies that the work was either (1) completed but at a level of achievement that is not worthy of credit or (2) was not completed and there was no agreement between the instructor and the student that the student would be awarded an I (see also I).
I Incomplete Assigned at the discretion of the instructor when, due to extraordinary circumstances, e.g., hospitalization, a student is prevented from completing the work of the course on time. Requires a written agreement between instructor and student.

Sexual Harrasment

“Sexual harassment” means unwelcome sexual advances, requests for sexual favors, and/or other verbal or physical conduct of a sexual nature. Such conduct has the purpose or effect of unreasonably interfering with an individual’s work or academic performance or creating an intimidating, hostile, or offensive working or academic environment in any University activity or program. Such behavior is not acceptable in the University setting. For additional information, please consult Board of Regents Policy: http://www1.umn.edu/regents/policies/humanresources/SexHarassment.html

Student Conduct Code

The University seeks an environment that promotes academic achievement and integrity, that is protective of free inquiry, and that serves the educational mission of the University. Similarly, the University seeks a community that is free from violence, threats, and intimidation; that is respectful of the rights, opportunities, and welfare of students, faculty, staff, and guests of the University; and that does not threaten the physical or mental health or safety of members of the University community.

As a student at the University you are expected adhere to Board of Regents Policy: Student Conduct Code. To review the Student Conduct Code, please see: http://regents.umn.edu/sites/default/files/policies/Student_Conduct_Code.pdf. Note that the conduct code specifically addresses disruptive classroom conduct.