Downloads & Free Reading Options - Results

Following the success of last year's teaching and career development workshop, this year's DART-P sponsored workshop at the Psychology Postgraduate Affairs Group (PsyPAG) Annual Conference held at Lancaster University focused on postgraduate's experiences of teaching research methods. This article provides a review of the invited speakers presentation, Dr. Emma McDonald, the Higher Education Academy (HEA) Psychology Technician of the year.

Following the success of last year's teaching and career development workshop, this year's DART-P sponsored workshop at the Psychology Postgraduate Affairs Group (PsyPAG) Annual Conference held at Lancaster University focused on postgraduate's experiences of teaching research methods. This article provides a review of the invited speakers presentation, Dr. Emma McDonald, the Higher Education Academy (HEA) Psychology Technician of the year.

Following the success of last year's teaching and career development workshop, this year's DART-P sponsored workshop at the Psychology Postgraduate Affairs Group (PsyPAG) Annual Conference held at Lancaster University focused on postgraduate's experiences of teaching research methods. This article provides a review of the invited speakers presentation, Dr. Emma McDonald, the Higher Education Academy (HEA) Psychology Technician of the year.

This study examined the relationship between students' scores on Research Methods and statistics, and undergraduate project at the final year. The purpose was to find out whether students matched knowledge of research with project-writing skill. The study adopted an expost facto correlational design. Scores on Research Methods and Statistics for 2008/2009, 2009/2010, 2010/2011 were correlated with project scores for 2009/2010, 2010/2011 and 2011/2012 respectively. The population for 2009/2010, 2010/2011 and 2011/2012 were 86, 90 and 102 respectively. The respective samples drawn through systematic random sampling technique were 23, 25 and 30. The past questions on Research Methods and Statistics for 2008/2009, 2009/2010, 2010/2011 as well as the undergraduate project assessment for 2009/2010, 2010/2011 and 2011/2012 were the instruments used for data collection. The three hypotheses for the study were tested at 0.05 level of significance using the Product Moment Correlation techniques. The co-efficient of determination was used to answer the research questions. The result of the study showed a significant relationship between students' scores on Research Methods and Statistics and undergraduate projects for 2009/2010 academic session only. The conclusion was that scores on Research Methods and Statistics did not adequately correlate with students' undergraduate project scores. It was recommended that emphasis on Research Methods and Statistics should be placed on the problem-solving nature of research.

This study examined the relationship between students' scores on Research Methods and statistics, and undergraduate project at the final year. The purpose was to find out whether students matched knowledge of research with project-writing skill. The study adopted an expost facto correlational design. Scores on Research Methods and Statistics for 2008/2009, 2009/2010, 2010/2011 were correlated with project scores for 2009/2010, 2010/2011 and 2011/2012 respectively. The population for 2009/2010, 2010/2011 and 2011/2012 were 86, 90 and 102 respectively. The respective samples drawn through systematic random sampling technique were 23, 25 and 30. The past questions on Research Methods and Statistics for 2008/2009, 2009/2010, 2010/2011 as well as the undergraduate project assessment for 2009/2010, 2010/2011 and 2011/2012 were the instruments used for data collection. The three hypotheses for the study were tested at 0.05 level of significance using the Product Moment Correlation techniques. The co-efficient of determination was used to answer the research questions. The result of the study showed a significant relationship between students' scores on Research Methods and Statistics and undergraduate projects for 2009/2010 academic session only. The conclusion was that scores on Research Methods and Statistics did not adequately correlate with students' undergraduate project scores. It was recommended that emphasis on Research Methods and Statistics should be placed on the problem-solving nature of research.

389 p. ; 24 cm

This report documents the National Center for Health Statistics' (NCHS) methods for bridging the Census 2000 multiple-race resident population to single-race categories and describing bridged race resident population estimates. Data came from the pooled 1997-2000 National Health Interview Surveys. The bridging models included demographic and contextual covariates at the person or county level. Allocation probabilities from the regression models were applied to the Census Bureau's 2000 Modified Race Data Summary File population counts to assign multiple-race persons to single-race categories. Bridging has the most impact on American Indian and Alaska Native (AIAN) and Asian or Pacific Islander (API) populations, a small impact on the black population, and a negligible impact on the white population. For the country overall, the AIAN, API black, and white bridged population counts are 12.0, 5.0, 2.5, and 0.5 percent higher than the corresponding Census 2000 single-race counts. There is considerably more variation at the subnational than the national level. Bridged single-race population counts have been used to calculate birth and death rates produced by NCHS for 2000 and 2001 and to revise previously published rates for the 1990s, 2000, and 2001. The bridging methodology will be used to bridge postcensal population estimates for later years. Bridged population counts presented here and in subsequent years may be updated as additional data become available for use in the bridging process. File layouts are appended. (SM)

This report documents the National Center for Health Statistics' (NCHS) methods for bridging the Census 2000 multiple-race resident population to single-race categories and describing bridged race resident population estimates. Data came from the pooled 1997-2000 National Health Interview Surveys. The bridging models included demographic and contextual covariates at the person or county level. Allocation probabilities from the regression models were applied to the Census Bureau's 2000 Modified Race Data Summary File population counts to assign multiple-race persons to single-race categories. Bridging has the most impact on American Indian and Alaska Native (AIAN) and Asian or Pacific Islander (API) populations, a small impact on the black population, and a negligible impact on the white population. For the country overall, the AIAN, API black, and white bridged population counts are 12.0, 5.0, 2.5, and 0.5 percent higher than the corresponding Census 2000 single-race counts. There is considerably more variation at the subnational than the national level. Bridged single-race population counts have been used to calculate birth and death rates produced by NCHS for 2000 and 2001 and to revise previously published rates for the 1990s, 2000, and 2001. The bridging methodology will be used to bridge postcensal population estimates for later years. Bridged population counts presented here and in subsequent years may be updated as additional data become available for use in the bridging process. File layouts are appended. (SM)

Structural equation modeling (SEM) has been deemed as a proper method when variables contain measurement errors. In contrast, path analysis with composite-scores is preferred for prediction and diagnosis of individuals. While path analysis with composite-scores has been criticized for yielding biased parameter estimates, recent literature pointed out that the population values of parameters in a latent-variable model depend on artificially assigned scales. Consequently, bias in parameter estimates is not a well-grounded concept for models involving latent constructs. This article compares path analysis with composite-scores against SEM with respect to effect size and statistical power in testing the significance of the path coefficients, via the z- or t-statistics. The data come from many sources with various models that are substantively determined. Results show that SEM is not as powerful as path analysis even with equally-weighted-composites. But path analysis with Bartlett-factor- scores and the partial-least-squares approach to SEM perform the best with respect to effect size and power. [This paper will be published in "Behavior Research Methods." Discrepancy between the title of the article and authored paper.]

Structural equation modeling (SEM) has been deemed as a proper method when variables contain measurement errors. In contrast, path analysis with composite-scores is preferred for prediction and diagnosis of individuals. While path analysis with composite-scores has been criticized for yielding biased parameter estimates, recent literature pointed out that the population values of parameters in a latent-variable model depend on artificially assigned scales. Consequently, bias in parameter estimates is not a well-grounded concept for models involving latent constructs. This article compares path analysis with composite-scores against SEM with respect to effect size and statistical power in testing the significance of the path coefficients, via the z- or t-statistics. The data come from many sources with various models that are substantively determined. Results show that SEM is not as powerful as path analysis even with equally-weighted-composites. But path analysis with Bartlett-factor- scores and the partial-least-squares approach to SEM perform the best with respect to effect size and power. [This paper will be published in "Behavior Research Methods." Discrepancy between the title of the article and authored paper.]

Structural equation modeling (SEM) has been deemed as a proper method when variables contain measurement errors. In contrast, path analysis with composite-scores is preferred for prediction and diagnosis of individuals. While path analysis with composite-scores has been criticized for yielding biased parameter estimates, recent literature pointed out that the population values of parameters in a latent-variable model depend on artificially assigned scales. Consequently, bias in parameter estimates is not a well-grounded concept for models involving latent constructs. This article compares path analysis with composite-scores against SEM with respect to effect size and statistical power in testing the significance of the path coefficients, via the z- or t-statistics. The data come from many sources with various models that are substantively determined. Results show that SEM is not as powerful as path analysis even with equally-weighted-composites. But path analysis with Bartlett-factor- scores and the partial-least-squares approach to SEM perform the best with respect to effect size and power. [This paper will be published in "Behavior Research Methods." Discrepancy between the title of the article and authored paper.]

Structural equation modeling (SEM) has been deemed as a proper method when variables contain measurement errors. In contrast, path analysis with composite-scores is preferred for prediction and diagnosis of individuals. While path analysis with composite-scores has been criticized for yielding biased parameter estimates, recent literature pointed out that the population values of parameters in a latent-variable model depend on artificially assigned scales. Consequently, bias in parameter estimates is not a well-grounded concept for models involving latent constructs. This article compares path analysis with composite-scores against SEM with respect to effect size and statistical power in testing the significance of the path coefficients, via the z- or t-statistics. The data come from many sources with various models that are substantively determined. Results show that SEM is not as powerful as path analysis even with equally-weighted-composites. But path analysis with Bartlett-factor- scores and the partial-least-squares approach to SEM perform the best with respect to effect size and power. [This paper will be published in "Behavior Research Methods." Discrepancy between the title of the article and authored paper.]

Structural equation modeling (SEM) has been deemed as a proper method when variables contain measurement errors. In contrast, path analysis with composite-scores is preferred for prediction and diagnosis of individuals. While path analysis with composite-scores has been criticized for yielding biased parameter estimates, recent literature pointed out that the population values of parameters in a latent-variable model depend on artificially assigned scales. Consequently, bias in parameter estimates is not a well-grounded concept for models involving latent constructs. This article compares path analysis with composite-scores against SEM with respect to effect size and statistical power in testing the significance of the path coefficients, via the z- or t-statistics. The data come from many sources with various models that are substantively determined. Results show that SEM is not as powerful as path analysis even with equally-weighted-composites. But path analysis with Bartlett-factor- scores and the partial-least-squares approach to SEM perform the best with respect to effect size and power. [This paper will be published in "Behavior Research Methods." Discrepancy between the title of the article and authored paper.]

Structural equation modeling (SEM) has been deemed as a proper method when variables contain measurement errors. In contrast, path analysis with composite-scores is preferred for prediction and diagnosis of individuals. While path analysis with composite-scores has been criticized for yielding biased parameter estimates, recent literature pointed out that the population values of parameters in a latent-variable model depend on artificially assigned scales. Consequently, bias in parameter estimates is not a well-grounded concept for models involving latent constructs. This article compares path analysis with composite-scores against SEM with respect to effect size and statistical power in testing the significance of the path coefficients, via the z- or t-statistics. The data come from many sources with various models that are substantively determined. Results show that SEM is not as powerful as path analysis even with equally-weighted-composites. But path analysis with Bartlett-factor- scores and the partial-least-squares approach to SEM perform the best with respect to effect size and power. [This paper will be published in "Behavior Research Methods." Discrepancy between the title of the article and authored paper.]

Structural equation modeling (SEM) has been deemed as a proper method when variables contain measurement errors. In contrast, path analysis with composite-scores is preferred for prediction and diagnosis of individuals. While path analysis with composite-scores has been criticized for yielding biased parameter estimates, recent literature pointed out that the population values of parameters in a latent-variable model depend on artificially assigned scales. Consequently, bias in parameter estimates is not a well-grounded concept for models involving latent constructs. This article compares path analysis with composite-scores against SEM with respect to effect size and statistical power in testing the significance of the path coefficients, via the z- or t-statistics. The data come from many sources with various models that are substantively determined. Results show that SEM is not as powerful as path analysis even with equally-weighted-composites. But path analysis with Bartlett-factor- scores and the partial-least-squares approach to SEM perform the best with respect to effect size and power. [This paper will be published in "Behavior Research Methods." Discrepancy between the title of the article and authored paper.]

Structural equation modeling (SEM) has been deemed as a proper method when variables contain measurement errors. In contrast, path analysis with composite-scores is preferred for prediction and diagnosis of individuals. While path analysis with composite-scores has been criticized for yielding biased parameter estimates, recent literature pointed out that the population values of parameters in a latent-variable model depend on artificially assigned scales. Consequently, bias in parameter estimates is not a well-grounded concept for models involving latent constructs. This article compares path analysis with composite-scores against SEM with respect to effect size and statistical power in testing the significance of the path coefficients, via the z- or t-statistics. The data come from many sources with various models that are substantively determined. Results show that SEM is not as powerful as path analysis even with equally-weighted-composites. But path analysis with Bartlett-factor- scores and the partial-least-squares approach to SEM perform the best with respect to effect size and power. [This paper will be published in "Behavior Research Methods." Discrepancy between the title of the article and authored paper.]

This report summarizes current knowledge and research on the quality and reliability of death rates by race and Hispanic origin in official mortality statistics of the United States produced by the National Center for Health Statistics (NCHS). It provides a quantitative assessment of bias in death rates by race and Hispanic origin and identifies areas for targeted research. Death rates are based on information on deaths from death certificates filed in the states and compiled into a national database by NCHS, and on population data from the Census Bureau. Researchers summarized selected studies of race/Hispanic origin misclassification and undercoverage on deaths and population. They made estimates on the separate and joint bias on death rates by race and Hispanic origin from the two sources. They made simplifying assumptions about the stability of the biases over time and among age groups. Original results are presented using an expanded, updated database from the National Longitudinal Mortality Study. Death rates for all groups show net effects of race misclassification and undercoverage. For whites and blacks, published death rates are overstated in official publications by 1.0 percent and 5.0 percent, respectively, resulting mainly from undercounts in the census. Death rates for other minority groups are understated as follows: American Indians, 21 percent; Asian or Pacific Islanders: 11 percent; and Hispanics, 2 percent. (SM)

This report summarizes current knowledge and research on the quality and reliability of death rates by race and Hispanic origin in official mortality statistics of the United States produced by the National Center for Health Statistics (NCHS). It provides a quantitative assessment of bias in death rates by race and Hispanic origin and identifies areas for targeted research. Death rates are based on information on deaths from death certificates filed in the states and compiled into a national database by NCHS, and on population data from the Census Bureau. Researchers summarized selected studies of race/Hispanic origin misclassification and undercoverage on deaths and population. They made estimates on the separate and joint bias on death rates by race and Hispanic origin from the two sources. They made simplifying assumptions about the stability of the biases over time and among age groups. Original results are presented using an expanded, updated database from the National Longitudinal Mortality Study. Death rates for all groups show net effects of race misclassification and undercoverage. For whites and blacks, published death rates are overstated in official publications by 1.0 percent and 5.0 percent, respectively, resulting mainly from undercounts in the census. Death rates for other minority groups are understated as follows: American Indians, 21 percent; Asian or Pacific Islanders: 11 percent; and Hispanics, 2 percent. (SM)

This report summarizes current knowledge and research on the quality and reliability of death rates by race and Hispanic origin in official mortality statistics of the United States produced by the National Center for Health Statistics (NCHS). It provides a quantitative assessment of bias in death rates by race and Hispanic origin and identifies areas for targeted research. Death rates are based on information on deaths from death certificates filed in the states and compiled into a national database by NCHS, and on population data from the Census Bureau. Researchers summarized selected studies of race/Hispanic origin misclassification and undercoverage on deaths and population. They made estimates on the separate and joint bias on death rates by race and Hispanic origin from the two sources. They made simplifying assumptions about the stability of the biases over time and among age groups. Original results are presented using an expanded, updated database from the National Longitudinal Mortality Study. Death rates for all groups show net effects of race misclassification and undercoverage. For whites and blacks, published death rates are overstated in official publications by 1.0 percent and 5.0 percent, respectively, resulting mainly from undercounts in the census. Death rates for other minority groups are understated as follows: American Indians, 21 percent; Asian or Pacific Islanders: 11 percent; and Hispanics, 2 percent. (SM)

This report summarizes current knowledge and research on the quality and reliability of death rates by race and Hispanic origin in official mortality statistics of the United States produced by the National Center for Health Statistics (NCHS). It provides a quantitative assessment of bias in death rates by race and Hispanic origin and identifies areas for targeted research. Death rates are based on information on deaths from death certificates filed in the states and compiled into a national database by NCHS, and on population data from the Census Bureau. Researchers summarized selected studies of race/Hispanic origin misclassification and undercoverage on deaths and population. They made estimates on the separate and joint bias on death rates by race and Hispanic origin from the two sources. They made simplifying assumptions about the stability of the biases over time and among age groups. Original results are presented using an expanded, updated database from the National Longitudinal Mortality Study. Death rates for all groups show net effects of race misclassification and undercoverage. For whites and blacks, published death rates are overstated in official publications by 1.0 percent and 5.0 percent, respectively, resulting mainly from undercounts in the census. Death rates for other minority groups are understated as follows: American Indians, 21 percent; Asian or Pacific Islanders: 11 percent; and Hispanics, 2 percent. (SM)

This report summarizes current knowledge and research on the quality and reliability of death rates by race and Hispanic origin in official mortality statistics of the United States produced by the National Center for Health Statistics (NCHS). It provides a quantitative assessment of bias in death rates by race and Hispanic origin and identifies areas for targeted research. Death rates are based on information on deaths from death certificates filed in the states and compiled into a national database by NCHS, and on population data from the Census Bureau. Researchers summarized selected studies of race/Hispanic origin misclassification and undercoverage on deaths and population. They made estimates on the separate and joint bias on death rates by race and Hispanic origin from the two sources. They made simplifying assumptions about the stability of the biases over time and among age groups. Original results are presented using an expanded, updated database from the National Longitudinal Mortality Study. Death rates for all groups show net effects of race misclassification and undercoverage. For whites and blacks, published death rates are overstated in official publications by 1.0 percent and 5.0 percent, respectively, resulting mainly from undercounts in the census. Death rates for other minority groups are understated as follows: American Indians, 21 percent; Asian or Pacific Islanders: 11 percent; and Hispanics, 2 percent. (SM)

This content analysis study evaluated the research methods and statistical techniques employed in studies reported in the professional journal, "TECSE" (Topics in Early Childhood Special Education) between 1990 and 1993. Research articles comprised 37 percent of all articles (n=120) published during this period. The most frequently utilized statistical techniques were descriptive statistics, t-tests, multiple regression, analysis of variance, and correlation. Nonintervention designs represented the majority of research articles, with causal-comparative being the most prominent type. Of the intervention designs, pretest/posttest was the most frequently utilized method. The majority of article authors were female. (DB)

This content analysis study evaluated the research methods and statistical techniques employed in studies reported in the professional journal, "TECSE" (Topics in Early Childhood Special Education) between 1990 and 1993. Research articles comprised 37 percent of all articles (n=120) published during this period. The most frequently utilized statistical techniques were descriptive statistics, t-tests, multiple regression, analysis of variance, and correlation. Nonintervention designs represented the majority of research articles, with causal-comparative being the most prominent type. Of the intervention designs, pretest/posttest was the most frequently utilized method. The majority of article authors were female. (DB)

Synthetic estimation is a statistical technique that estimates small-area statistics by combining national estimates of the relevant characteristics with estimates of other known characteristics of the small geographic area. The advantages of the synthetic estimation approach to local estimation are its intuitive appeal, its simplicity, and its low cost. A major disadvantage is its possibile lack of sensitivity to certain local characteristics. Another method used for the same purpose is the "nearly unbiased" estimator. The mathematics of both methods are presented, including formulas for estimating mean square error and average mean square error. An evaluation of synthetic estimates is demonstrated by comparing the results of a 50-cell grid with the results from 2, 4, 8, and 16-cell grids. (Author/CTM)

Synthetic estimation is a statistical technique that estimates small-area statistics by combining national estimates of the relevant characteristics with estimates of other known characteristics of the small geographic area. The advantages of the synthetic estimation approach to local estimation are its intuitive appeal, its simplicity, and its low cost. A major disadvantage is its possibile lack of sensitivity to certain local characteristics. Another method used for the same purpose is the "nearly unbiased" estimator. The mathematics of both methods are presented, including formulas for estimating mean square error and average mean square error. An evaluation of synthetic estimates is demonstrated by comparing the results of a 50-cell grid with the results from 2, 4, 8, and 16-cell grids. (Author/CTM)

Feedback from students on a Level 1 Research Methods and Statistics module, studied as a core part of a BSc Psychology programme, highlighted demand for additional tutorials to help them to understand basic concepts. Students in their final year of study commonly request work experience to enhance their employability. All students on the Level 1 Research Methods and Statistics module were invited to sign up for PAL (Peer-Assisted Learning) sessions which offered interactive learning under the guidance of a PAL Student-Facilitator. Five students at undergraduate Level 3 and Masters level were recruited and provided with training to facilitate small group sessions. Session provision also included materials for interactive exercises. A wrap-up session considered how Facilitators could draw on this experience to enhance their employability. Analysis of feedback from the Level 1 students showed their experience was uniformly good with sessions found to be useful in aiding understanding. Any student-learner dissatisfaction resulted from a continuing inability to feel confident with the basic concepts taught on the module, although students said their understanding had improved overall. Facilitators found the sessions to be rewarding and felt they had developed their interpersonal skills, their leadership skills, and their understanding of the subject matter, and enhanced their employability, particularly in terms of skills transference. Attendance from student-learners was poor compared to the number who signed up, but the small size of many groups enabled individual support which was expressly appreciated. The greatest student-learner demand was for sessions on topics directly related to coursework. Future PAL sessions will be directly related to these topics and arranged for the middle section of the semester, that is, when the Level 1 students have had time to identify their difficulties and the session can still be of maximum benefit for general learning and assessment completion. Considerations for student development and employability are also discussed. (Contains 1 figure and 1 table.)

The 26th Symposium on the Interface of Computing Science and Statistics was held on June 15-18, 1994 at the Sheraton Imperial Hotel in Research Triangle Park, NC. The conference theme was Computationally intensive Statistical Methods. The theme is especially appropriate as computational power has increased dramatically in the last few years and the use of resampling techniques has boomed. The Interface was scheduled between two other statistics conferences in the same area: the Spring Research Conference on Statistics in Industry, hosted by the National Institute of Statistical Sciences, and the Third World Congress-Bernoulli Society--IMS meetings, at the University of North Carolina in Chapel Hill. The conference attracted 363 attendees. There were 23 invited sessions, 21 contributed paper sessions, 9 poster presentations, 4 short courses, 2 practical tutorials, several statistical tutorial sessions, one keynote speech, one banquet presentation, and 2 tours. (AN)

The 26th Symposium on the Interface of Computing Science and Statistics was held on June 15-18, 1994 at the Sheraton Imperial Hotel in Research Triangle Park, NC. The conference theme was Computationally intensive Statistical Methods. The theme is especially appropriate as computational power has increased dramatically in the last few years and the use of resampling techniques has boomed. The Interface was scheduled between two other statistics conferences in the same area: the Spring Research Conference on Statistics in Industry, hosted by the National Institute of Statistical Sciences, and the Third World Congress-Bernoulli Society--IMS meetings, at the University of North Carolina in Chapel Hill. The conference attracted 363 attendees. There were 23 invited sessions, 21 contributed paper sessions, 9 poster presentations, 4 short courses, 2 practical tutorials, several statistical tutorial sessions, one keynote speech, one banquet presentation, and 2 tours. (AN)

The 26th Symposium on the Interface of Computing Science and Statistics was held on June 15-18, 1994 at the Sheraton Imperial Hotel in Research Triangle Park, NC. The conference theme was Computationally intensive Statistical Methods. The theme is especially appropriate as computational power has increased dramatically in the last few years and the use of resampling techniques has boomed. The Interface was scheduled between two other statistics conferences in the same area: the Spring Research Conference on Statistics in Industry, hosted by the National Institute of Statistical Sciences, and the Third World Congress-Bernoulli Society--IMS meetings, at the University of North Carolina in Chapel Hill. The conference attracted 363 attendees. There were 23 invited sessions, 21 contributed paper sessions, 9 poster presentations, 4 short courses, 2 practical tutorials, several statistical tutorial sessions, one keynote speech, one banquet presentation, and 2 tours. (AN)

The 26th Symposium on the Interface of Computing Science and Statistics was held on June 15-18, 1994 at the Sheraton Imperial Hotel in Research Triangle Park, NC. The conference theme was Computationally intensive Statistical Methods. The theme is especially appropriate as computational power has increased dramatically in the last few years and the use of resampling techniques has boomed. The Interface was scheduled between two other statistics conferences in the same area: the Spring Research Conference on Statistics in Industry, hosted by the National Institute of Statistical Sciences, and the Third World Congress-Bernoulli Society--IMS meetings, at the University of North Carolina in Chapel Hill. The conference attracted 363 attendees. There were 23 invited sessions, 21 contributed paper sessions, 9 poster presentations, 4 short courses, 2 practical tutorials, several statistical tutorial sessions, one keynote speech, one banquet presentation, and 2 tours. (AN)

The 26th Symposium on the Interface of Computing Science and Statistics was held on June 15-18, 1994 at the Sheraton Imperial Hotel in Research Triangle Park, NC. The conference theme was Computationally intensive Statistical Methods. The theme is especially appropriate as computational power has increased dramatically in the last few years and the use of resampling techniques has boomed. The Interface was scheduled between two other statistics conferences in the same area: the Spring Research Conference on Statistics in Industry, hosted by the National Institute of Statistical Sciences, and the Third World Congress-Bernoulli Society--IMS meetings, at the University of North Carolina in Chapel Hill. The conference attracted 363 attendees. There were 23 invited sessions, 21 contributed paper sessions, 9 poster presentations, 4 short courses, 2 practical tutorials, several statistical tutorial sessions, one keynote speech, one banquet presentation, and 2 tours. (AN)

This paper describes the conceptual base for the development of a computer-based expert system. After reviewing developments in computer-based learning and experiments with computer-assisted learning in statistics, the paper describes the nature of expert systems and desired attributes of expert systems in statistics. An overview of proposed research projects to develop a computer-based expert system research outliner/statistical tutor is presented. Current progress, anticipated timelines and methodological concerns are provided. Two figures--The Changing Focus of Attention in Technology for Computer-Assisted Learning and System Delivery Tools--are included. (Contains 37 references.) (Author/ALF)

This paper describes the conceptual base for the development of a computer-based expert system. After reviewing developments in computer-based learning and experiments with computer-assisted learning in statistics, the paper describes the nature of expert systems and desired attributes of expert systems in statistics. An overview of proposed research projects to develop a computer-based expert system research outliner/statistical tutor is presented. Current progress, anticipated timelines and methodological concerns are provided. Two figures--The Changing Focus of Attention in Technology for Computer-Assisted Learning and System Delivery Tools--are included. (Contains 37 references.) (Author/ALF)

This paper describes the conceptual base for the development of a computer-based expert system. After reviewing developments in computer-based learning and experiments with computer-assisted learning in statistics, the paper describes the nature of expert systems and desired attributes of expert systems in statistics. An overview of proposed research projects to develop a computer-based expert system research outliner/statistical tutor is presented. Current progress, anticipated timelines and methodological concerns are provided. Two figures--The Changing Focus of Attention in Technology for Computer-Assisted Learning and System Delivery Tools--are included. (Contains 37 references.) (Author/ALF)

The booklet presents a report on the operation and use of person numbering (PN) systems for vital and health statistics in Israel and Scandinavia. A PN system involves assigning a different number to each individual within a given society and using that number for commercial, governmental, and military identification. The purpose of the report is to provide information about PN systems to people involved in the public policy debate over whether such a system should be adopted in the United States. The document is presented in eight chapters. Chapter I explains how to develop a PN system and relates that Israel and the Scandinavian nations were selected as case studies because they are forerunners in developing and applying PN systems in national statistical programs. Chapters II through V focus on the PN systems of Sweden, Norway, Denmark, and Israel, respectively. Topics include organizing and operating the systems, problems peculiar to each country, administrative and statistical uses of the PN system, future use plans, and plans for additional studies and research. Chapter VI reviews advantages and limitations of PN systems. Advantages include that PN systems are technically easy to use and store and that they increase data available for statistics, planning, and research. Disadvantages include that the digits become reversed fairly frequently, number files are not always transferred when a person moves, and integration with other systems is difficult. Chapter VII focuses on issues of confidentiality and privacy. The final chapter presents an overview of PN systems in 16 nations, including the Netherlands, Peru, Australia, and Japan. (DB)

The report describes the initial design and the preliminary background exploration, subsequent development, and feasibility testing of methods for conducting a continuing National Ambulatory Medical Care Survey (NAMCS). Selected feasibility study findings are presented to illustrate collected data and suggest kinds of information that may be expected when substantive survey results become available on a continuing national basis. NAMCS was authorized to gather and disseminate statistical information about ambulatory health care provided by office-based physicians to the population of the United States. Ambulatory health care is defined as health services rendered individuals under their own cognizance, at a time when they are not in a hospital or other health care institution. Thirteen pages of detailed tables present the survey results regarding annual volume (1967-1972) and rates of patient visits for population groups, medical specialty groups, and geographic areas. Quantitative descriptions of visit characteristics include tabulations of patient's problems, reasons for visit, diagnoses, services, treatment, and subsequent disposition. The 40-page appendix presents the data collection forms for the survey and field tests. (AG)

In several studies of experimental interviewing techniques and their effect on reporting behavior described in this publication, an attempt is made to identify the elements of the interview process that are potential sources for improving data collection. Methodological studies designed to test the effectiveness of certain questionnaire designs and interviewing techniques used in the collection of data on health events in household interviews are presented. The role of behaviors, attitudes, perceptions, and information levels of both the respondent and the interviewer is investigated. (RC)

In several studies of experimental interviewing techniques and their effect on reporting behavior described in this publication, an attempt is made to identify the elements of the interview process that are potential sources for improving data collection. Methodological studies designed to test the effectiveness of certain questionnaire designs and interviewing techniques used in the collection of data on health events in household interviews are presented. The role of behaviors, attitudes, perceptions, and information levels of both the respondent and the interviewer is investigated. (RC)

In several studies of experimental interviewing techniques and their effect on reporting behavior described in this publication, an attempt is made to identify the elements of the interview process that are potential sources for improving data collection. Methodological studies designed to test the effectiveness of certain questionnaire designs and interviewing techniques used in the collection of data on health events in household interviews are presented. The role of behaviors, attitudes, perceptions, and information levels of both the respondent and the interviewer is investigated. (RC)

In several studies of experimental interviewing techniques and their effect on reporting behavior described in this publication, an attempt is made to identify the elements of the interview process that are potential sources for improving data collection. Methodological studies designed to test the effectiveness of certain questionnaire designs and interviewing techniques used in the collection of data on health events in household interviews are presented. The role of behaviors, attitudes, perceptions, and information levels of both the respondent and the interviewer is investigated. (RC)

In several studies of experimental interviewing techniques and their effect on reporting behavior described in this publication, an attempt is made to identify the elements of the interview process that are potential sources for improving data collection. Methodological studies designed to test the effectiveness of certain questionnaire designs and interviewing techniques used in the collection of data on health events in household interviews are presented. The role of behaviors, attitudes, perceptions, and information levels of both the respondent and the interviewer is investigated. (RC)

In several studies of experimental interviewing techniques and their effect on reporting behavior described in this publication, an attempt is made to identify the elements of the interview process that are potential sources for improving data collection. Methodological studies designed to test the effectiveness of certain questionnaire designs and interviewing techniques used in the collection of data on health events in household interviews are presented. The role of behaviors, attitudes, perceptions, and information levels of both the respondent and the interviewer is investigated. (RC)

In several studies of experimental interviewing techniques and their effect on reporting behavior described in this publication, an attempt is made to identify the elements of the interview process that are potential sources for improving data collection. Methodological studies designed to test the effectiveness of certain questionnaire designs and interviewing techniques used in the collection of data on health events in household interviews are presented. The role of behaviors, attitudes, perceptions, and information levels of both the respondent and the interviewer is investigated. (RC)

Psychology Books

Although the Intellectual Development (ID) index was constructed using standard psychometric procedures, the derivation of the other two indexes, Socio Intellectual Status (SIS) and Differential Intellectual Development (DID), by criterion scaling should have applications in diverse areas of scale or index construction. The ID is basically comparable to an intelligence quotient index. The SIS is somewhat analagous to a Socio-Economic Status (SES) index, while the DID reflects intellectual achievement independent of the SIS family background contribution to the ID Index. The use of the three indexes in studying findings from the Cycle II children's survey (aged 6-11 years) of 1963-65 helps to clarify some persistent issues related to intellectual achievement. The first-order analyses revealed statistically significant relationships between the ID Index and (1) number of pregnancies previous to the birth of the examined child; (2) twin versus nontwin birth status; and (3) attendance vs. nonattendance at nursery school and/or kindergarten. However, these relationships were mostly attributed to family background factors reflected in the Socio-Intellectual index. Although the numerical index values derived in this report are specific and limited to the data of the Health Examination Survey Cycle II (children) and Cycle III (youths), the data base should have widespread applicability to behavioral scientists. Statistical data are appended, as well as brief notes on criterion scaling, multiple linear regression, and skewness and kurtosis tests. (Author/CP)

Although the Intellectual Development (ID) index was constructed using standard psychometric procedures, the derivation of the other two indexes, Socio Intellectual Status (SIS) and Differential Intellectual Development (DID), by criterion scaling should have applications in diverse areas of scale or index construction. The ID is basically comparable to an intelligence quotient index. The SIS is somewhat analagous to a Socio-Economic Status (SES) index, while the DID reflects intellectual achievement independent of the SIS family background contribution to the ID Index. The use of the three indexes in studying findings from the Cycle II children's survey (aged 6-11 years) of 1963-65 helps to clarify some persistent issues related to intellectual achievement. The first-order analyses revealed statistically significant relationships between the ID Index and (1) number of pregnancies previous to the birth of the examined child; (2) twin versus nontwin birth status; and (3) attendance vs. nonattendance at nursery school and/or kindergarten. However, these relationships were mostly attributed to family background factors reflected in the Socio-Intellectual index. Although the numerical index values derived in this report are specific and limited to the data of the Health Examination Survey Cycle II (children) and Cycle III (youths), the data base should have widespread applicability to behavioral scientists. Statistical data are appended, as well as brief notes on criterion scaling, multiple linear regression, and skewness and kurtosis tests. (Author/CP)