Statistical simulation on microcomputers

Bradley, Drake R.; Senko, Michael W.; Stewart, Foster A.

doi:10.3758/bf03203153

Cited by 9 publications

(11 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sampling experiments like those reported in Table 4 can provide reasonably accurate estimates of these effects, provided that the number of simulated datasets employed in the simulation is large. As noted elsewhere (Bradley et al, 1990, p. 244), DATASIM estimates of Type I error rates have, in several cases, been more accurate than the "exact" values reported in previous analytical studies (Box, 1954;Hsu, 1938). So, the fact that DATASIM estimates are not exact is less of a concern than might seem the case on first consideration.…”

Section: Adjusting the Critical Value For Violationsmentioning

confidence: 73%

“…DATASIM processed in various ways. For example, the number of Type I or Type IT errors can be tabulated, the sampling distributions of test statistics can be plotted and compared to theoretical distributions using the Kolmogorov-Smimov test, and so on (Bradley et al, 1990). These capabilities are illustrated in subsequent sections of this paper.…”

Section: Generating Simulated Datamentioning

confidence: 99%

“…By storing the relevant commands in a file, the instructor can quickly initialize a simulation of a particular study, generate a simulated dataset, and display and analyze the results. Parameters of the simulation can be easily changed, and new datasets generated and displayed in order to assess what effects, if any, the changes have produced (see Bradley, 1989aBradley, , 1989bBradley, Senko, & Stewart, 1990).…”

mentioning

confidence: 99%

See 2 more Smart Citations

Anatomy of a DATASIM simulation: The Doob and Gross horn-honking study

Bradley

1991

Behavior Research Methods, Instruments, & Computers

Self Cite

View full text Add to dashboard Cite

The program DATASIM is used to simulate the classic "horn-honking" study by Doob and Gross (1968). In a 2 x 2 field experiment, Doob and Gross investigated the effects of status of frustrator-a low-or high-status car blocking an intersection-on the latency to honk among male and female drivers. The present paper illustrates how to extract the values of simulation parameters from the published study, how to initialize the simulation in DATASIM, and how to generate and analyze the simulated data. Certain complications arise because the latency data collected by Doob and Gross were nonnormally distributed, cell variances were heterogeneous, and sample sizes were unequal. DATASIM is able to incorporate these features in the simulation, and several methods for assessing the quality ofthe simulation are illustrated. In addition, sampling experiments are reported, which were performed to assess the joint and individual effects of nonnormality and heterogeneity on the Type I and Type II error rates of the F test. The paper concludes with some practical suggestions regarding how researchers can evaluate, and adjust for, the effects of such violations. 190In 1968, Doob and Gross published a study in which they investigated the effects of status on inhibiting aggression. The experimental paradigm was simple: The experimenters would drive a low-or high-status car to an intersection, wait for the light to change to green, and then just remain stationary. They were interested in measuring the hom-honking behavior of the driver in the car behind them, which they did by counting the total number of honks made during the 12-sec cycle of the light, and by ascertaining the latency to the first honk. (If the driver never honked, he or she was assigned the maximum possible latency of 12 sec.) The experimenters recorded the sex of the driver, and subsequently analyzed the latency data by conducting a 2 x2 (status x sex) analysis of variance.

show abstract

Section: Adjusting the Critical Value For Violationsmentioning

confidence: 73%

Section: Generating Simulated Datamentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Anatomy of a DATASIM simulation: The Doob and Gross horn-honking study

Bradley

1991

Behavior Research Methods, Instruments, & Computers

Self Cite

View full text Add to dashboard Cite

show abstract

“…DataSim (Bradley, Senko,& Stewart, 1990) is also recommended as a general-purpose statistical package for univariate data analysis. This package differs from the two previous programs because it is designed to conduct sampling experiments and will generate simulated data.…”

Section: Discussionmentioning

confidence: 99%

A review of macintosh-based instructional software

Goolkasian

1993

Behavior Research Methods, Instruments, & Computers

View full text Add to dashboard Cite

This paper reviews Macintosh-based instructional software. The programs that are discussed can be used as laboratory exercises for data collection and data analysis, or for computer-assisted instruction. Some suggestions for state-of-the-art software in each of these categories are made along with some comments regarding instructional objectives for each set of programs.Instructional software has changed significantly since the early 1980s when computer-based laboratories were first introduced (Goolkasian, 1985). We have recently upgraded our undergraduate computer laboratory by installing an AppleTalk network with Macintosh IIsi computers and have some suggestions for state-of-the-art instructional software. This review follows in the tradition of Hornby and Anderson (1990) and Butler and Neudecker (1989) by providing some comparative data and, also, updates previous work with a state-of-the-art library of programs for psychology instruction.The programs are compatible with System 7, and run effectively on a Macintosh with a coprocessor, at least 5 MB of memory and an 8Q-MB hard drive. The laboratory functions as a student center for data collection and data analysis, and each piece of software was chosen to meet specific instructional goals. Although the laboratory was developed primarily for students enrolled in Research Methods in Psychology, it is also used for demonstrations in upper division courses (i.e., Sensation and Perception, Physiological Psychology, and Human Cognitive Processes) and by students enrolled in independent research. Table 1 lists suggested software in each of three categories. Laboratory ExercisesThe programs in the first category provide the means by which students can gain "hands-on" experience in dataThis work was supported in part by funds from the foundation of the University of North Carolina at Charlotte and the State of North Carolina. Funds for the equipment to upgrade the Research Methods Laboratory was provided from National Science Foundation Grant USE-9151966. Correspondence should be addressed to P. Goolkasian, Department of Psychology, University of North Carolina, Charlotte, NC 28223 (e-mail: fpyOOpag@unccvm.bitnet).collection with a wide variety of experimental procedures. Although all of the programs in this category are used for data collection, they differ in the extent to which a laboratory procedure can be modified. Some provide ready-torun exercises (i.e., Experiments in Cognitive Psychology) that follow a contemporary paradigm from the psychologicalliterature while others (i.e., SuperLab) are tool-based and require the user to set up a customized experiment by using a series of interactive menus. The ease with which the software can be modified to customize the laboratory procedure can be an important variable when one is trying to meet specific instructional objectives. For example, ready-to-run programs are useful for providing students with a data-gathering experience in a specific area or with a specific research design. Experiments in Cognitive Psychology,...

show abstract

“…Moreover, complete data sets can be generated for a variety of experimental, multivariate, and contingency table designs. Previous publications have described the basic design and capabilities of the software (Bradley, 1989a(Bradley, , 1989b, evaluated the accuracy of the random number algorithms (Bradley, Senko, & Stewart, 1990), and reviewed the methodology for conducting a simulation of a previously published study (Bradley, 1991a).…”

mentioning

confidence: 99%

A simulation laboratory for statistics

Bradley

Hemstreet

Ziegenhagen

1992

Behavior Research Methods, Instruments, & Computers

Self Cite

View full text Add to dashboard Cite

In this paper, we describe a general purpose data simulator, Datasim, which is useful for anyone conducting computer-based laboratory assignments in statistics. Simulations illustrating sampling distributions, the central limit theorem, Type I and Type II decision errors, the power of a test, the effects of violating assumptions, and the distinction between orthogonal and nonorthogonal contrasts are discussed. Simulations illustrating other statistical concepts-partial correlation, regression to the mean, heteroscedasticity, the partitioning of error terms in splitplot designs, and so on-ean be developed easily. Simulations can be assigned as laboratory exercises, or the instructor can execute the simulations during class, integrate the results into an ongoing lecture, and use the results to initiate class discussion of the relevant statistical concepts.Over the last several decades, instructors of statistics have experimented with the use of simulation to communicate principles of probability, sampling theory, and power analysis. Although innovative in spirit, these efforts have been hampered by the lack of a convenient method for conducting sampling experiments.1 This paper reviews a software package, called Datasim, developed specifically for this purpose (Bradley, 1988(Bradley, , 1991b. In contrast to the "coin-flipping" and "dice-rolling" labs of a previous era, Datasim is a relatively sophisticated data generator: data can be generated from discrete or continuous probability distributions, from normal or nonnormal distributions, and from populations having correlated or uncorrelated observations. Moreover, complete data sets can be generated for a variety of experimental, multivariate, and contingency table designs. Previous publications have described the basic design and capabilities of the software (Bradley, 1989a(Bradley, , 1989b, evaluated the accuracy of the random number algorithms (Bradley, Senko, & Stewart, 1990), and reviewed the methodology for conducting a simulation of a previously published study (Bradley, 1991a).Although Datasim has a number of useful applications.! the present paper focuses exclusively on computer-based laboratory assignments in which simulation is used to demonstrate one or more key concepts in statistics. The first section of the paper introduces the way in which one initializes a simulation and generates single and multiple data sets. Subsequent sections apply this method to laboratory simulationsdemonstrating the central limit theorem, Type I and Type II decision errors, the effects of violating assumptions, and orthogonal and nonorthogonal contrasts.The software described in this paper was developed by the author and is available at nominal cost from Desktop Press, 90 Bardwell 51., Lewiston, ME 04240 (phone: 207-786-4113 or 6180). Requests for reprints of this article should beaddressed to Drake R. Bradley, Department of Psychology, Bates College, Lewiston, ME 04240. Generating Simulated DataThe first step in generating simulated data is to initialize the simulation....

show abstract

Statistical simulation on microcomputers

Cited by 9 publications

References 16 publications

Anatomy of a DATASIM simulation: The Doob and Gross horn-honking study

Anatomy of a DATASIM simulation: The Doob and Gross horn-honking study

A review of macintosh-based instructional software

A simulation laboratory for statistics

Contact Info

Product

Resources

About