Integrative data analysis: The simultaneous analysis of multiple data sets.

There are both quantitative and methodological techniques that foster the development and maintenance of a cumulative knowledge base within the psychological sciences. Most noteworthy of these techniques is meta-analysis, which allows for the synthesis of summary statistics drawn from multiple studies when the original data are not available. However, when the original data can be obtained from multiple studies, many advantages stem from the statistical analysis of the pooled data. The authors define integrative data analysis (IDA) as the analysis of multiple data sets that have been pooled into one. Although variants of IDA have been incorporated into other scientific disciplines, the use of these techniques is much less evident in psychology. In this article the authors present an overview of IDA as it may be applied within the psychological sciences, discuss the relative advantages and disadvantages of IDA, describe analytic strategies for analyzing pooled individual data, and offer recommendations for the use of IDA in practice. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

浅谈统计数据质量

近年来,我国经济持续、快速增长,引起了世界的关注,而作为衡量经济发展规模和水平的统计数据及其质量也成为关注的焦点。为了更好地满足社会经济发展过程中对统计数据的需求,要加强统计数据质量管理、提高数据质量,必须加强统计管理体制,加强监督职能,建立健全的数据质量评估机制。     

The problem of inference from curves based on group data.

The use of curves based on averaged data to infer the nature of individual curves or functional relationships is hazardous only when interpretations of the group data, or inferences derived from them, are unwarranted and violate accepted principles of statistical inference. The problems involved in and the procedures appropriate to each of 3 mathematical functions are discussed: Class A, Functions unmodified by averaging; Class B, Functions for which averaging complicates the interpretation of parameters but leaves form unchanged; and Class C, Functions modified in form by averaging. The form of a group mean curve may provide a way to test exact hypotheses about individual curves, although the form of the latter is not determined by the form of the group mean curve. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Pooling data from multiple longitudinal studies: The role of item response theory in integrative data analysis.

There are a number of significant challenges researchers encounter when studying development over an extended period of time, including subject attrition, the changing of measurement structures across groups and developmental periods, and the need to invest substantial time and money. Integrative data analysis is an emerging set of methodologies that allows researchers to overcome many of the challenges of single-sample designs through the pooling of data drawn from multiple existing developmental studies. This approach is characterized by a host of advantages, but this also introduces several new complexities that must be addressed prior to broad adoption by developmental researchers. In this article, the authors focus on methods for fitting measurement models and creating scale scores using data drawn from multiple longitudinal studies. The authors present findings from the analysis of repeated measures of internalizing symptomatology that were pooled from three existing developmental studies. The authors describe and demonstrate each step in the analysis and conclude with a discussion of potential limitations and directions for future research. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Mixed-effects logistic regression for estimating transitional probabilities in sequentially coded observational data.

This article demonstrates the use of mixed-effects logistic regression (MLR) for conducting sequential analyses of binary observational data. MLR is a special case of the mixed-effects logit modeling framework, which may be applied to multicategorical observational data. The MLR approach is motivated in part by G. A. Dagne, G. W. Howe, C. H. Brown, & B. O. Muthén (2002) advances in general linear mixed models for sequential analyses of observational data in the form of contingency table frequency counts. The advantage of the MLR approach is that it circumvents obstacles in the estimation of random sampling error encountered using Dagne and colleagues' approach. This article demonstrates the MLR model in an analysis of observed sequences of communication in a sample of young adult same-sex peer dyads. The results obtained using MLR are compared with those of a parallel analysis using Dagne and colleagues' linear mixed model for binary observational data in the form of log odds ratios. Similarities and differences between the results of the 2 approaches are discussed. Implications for the use of linear mixed models versus mixed-effects logit models for sequential analyses are considered. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

SPSS统计软件包在矿山统计分析中的应用

文中以某矿山应用SPSS统计软件包为例 ,对该矿的技术经济指标进行数据分析 ,并对矿石品位数据进行统计检验。     

The analysis of profile data.

"Methods were suggested for handling 3 problems in the analysis of test profiles: measuring the similarity of profiles, discriminating the typical profiles of two or more groups, and clustering profiles into homogeneous groups. The suggested methods were, respectively, picturing profiles as interpoint distances in Euclidean space, use of the linear multiple-discriminant function, and factor analysis of profile cross-product terms. Some suggestions were given about transformations of profile data before further analysis." (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Introduction to functional data analysis.

Psychologists and behavioural scientists are increasingly collecting data that are drawn from continuous underlying processes. We describe a set of quantitative methods, Functional Data Analysis (FDA), which can answer a number of questions that traditional statistical approaches cannot. These methods are applicable for analyzing many datasets that are common in experimental psychology, including time series data, repeated measures, and data distributed over time or space as in neuroimaging experiments. The primary advantage of FDA is that it allows the researcher to ask questions about when in a time series differences may exist between two or more sets of observations. We discuss functional correlations, principal components, the derivatives of functional curves, and analysis of variances models. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Analytic methods for modeling longitudinal data from rolling therapy groups with membership turnover.

Interventions for a variety of emotional and behavioral problems are commonly delivered in the context of treatment groups, with many using rolling admission to sustain membership (i.e., admission, dropout, and discharge from group are perpetual and ongoing). The authors present an overview of the analytic challenges inherent in rolling group data and outline commonly used (but flawed) analytic and design approaches to addressing (or sidestepping) these issues. Moreover, the authors propose use of latent class pattern mixture models (LCPMMs) as a statistically and conceptually defensible approach for modeling treatment data from rolling groups. The LCPMM approach is illustrated with rolling group data from a group-based alcoholism pilot treatment trial (N = 128). Different inferences were made with regard to treatment efficacy under LCPMM vs. the commonly used standard group-clustered latent growth model (LGM); coupled with other preliminary findings in this area, inferences from LGMs may be overly liberal when applied to data from rolling groups. Continued work on data analytic difficulties in groups with membership turnover is critical for furthering the ecological validity of research on behavioral treatments. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

On the equivalence of clinical and statistical methods.

Based on material from a doctoral thesis, an attempt was made to resolve an inconsistency in Meehl's analysis of statistical vs. clinical methods of assessment (see 29: 3991). "Eight interviewers assessed from 14 to 50 Canadian Army applicants using information obtained from biographical and test data, and from interview conversation. Each applicant was described on a 120-item Q-sort check list. These data were quantified and combined into composite statistical scores (biographical and test data) and clinical scores (Q-sort data). The results indicated that: (a) clinical scores were associated more closely with decisions than were statistical scores; (b) statistical and clinical scores correlated at a low level; (c) the decisions of different interviewers were associated with the same Q-sort, biographical, and test data; and (d) statistical and clinical scores were comparable in reliability." "… this study… strengthens the status of the clinical assessment." (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

On the documentation of statistical analyses in the "clicky-box" era.

The proliferation of user-friendly computer packages that perform statistical analyses has led to what we call the "clicky-box" era in statistics: It is now possible to run complicated statistical analyses by clicking on menu options and reading the output. Although the advent of such technology has had a profoundly positive impact on psychological research (think of doing a factor analysis or fitting a structural equation model without such software!), it has also led to a greater opportunity for confusion. This article provides some essential procedures for the documentation and safe handling of data. The procedures outlined serve as a guide to help avoid the problems and confusions that can easily arise in the clicky-box era of highly automated data analysis. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Some factors in the selection of leaders by members of small groups.

The present study was designed to investigate the relationship between amount of participation, frequency of task- and group-oriented interaction, and the selection of leaders by other group members. The general hypothesis was that amount of participation and organizational-integrative interaction are both associated with leader selection but that each may reflect different aspects of the criterion. Two three-person groups met simultaneously in separate rooms and were given identical instructions by the experimenters. Each group discussed a human relations problem for about 20 minutes, reaching a group consensus in that time. The groups were then told that a second meeting would take place immediately in which one person from each group would act as a representative; the two representatives would discuss a concrete problem related to the previously discussed topic. Each group then selected one of the members to act as representative. Data were obtained on 26 three-person groups, and for the total sample of 66 Ss, data on time-talked were put into percentage form for each meeting. Those members who were chosen as representatives talked an average of 44.8% of the meeting time; nonrepresentatives talked an average of 27.6% of the time. The average number of interactions (DEF) per meeting was 12.6 for representatives and 6.1 for nonrepresentatives. However, the variances are heterogeneous in this case, so a median test was used, yielding a chi square of 8.79, significant beyond the .01 level. Thus, both on measures of amount of participation and interaction, those members chosen as representatives were higher. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Reporting results of latent growth modeling and multilevel modeling analyses: Some recommendations for rehabilitation psychology.

Objective: There has been a general increase in interest and use of modeling techniques that treat data as nested, whether it is people nested within larger units, such as families or treatment centers, or observations nested under people. The popularity can be witnessed by noting the number of new textbooks and articles related to latent growth curve modeling and multilevel modeling. This paper discusses both of these techniques in the context of longitudinal research designs, with the main purposes of highlighting some benefits and issues related to the use of these models and outlining guidelines for reporting results from studies using multilevel modeling or latent growth modeling. Implications: These longitudinal analytic techniques can be greatly beneficial to researchers conducting rehabilitation studies, but there are several issues related to their use and reporting that need to be taken into consideration. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The role of small groups in mediating the effects of propaganda.

The attitude toward wiretapping was measured for a group of students before and after exposure to a speech opposing their attitude. They were then asked to list on a choice questionnaire their first, second, and third choices people present, whose views were made public, whom they would like to listen to or speak with on the issue. The entire group then engaged in free discussion of the issue, after which the opinion questionnaire was again filled out. The results indicated that individuals whose confidence in a belief has been shaken by exposure to opposing propaganda, prefer to hear arguments from their own side to bolster their confidence and tend to listen preferentially with persons who agree with them, with the consequence that their confidence in their opinion soon returns to the original level. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Assumptions of cross-level measurement and structural invariance in the analysis of multilevel data: Problems and solutions.

This article demonstrates assumptions of invariance that researchers often implicitly make when analyzing multilevel data. The first set of assumptions is measurement-based and corresponds to the fact that researchers often conduct single-level exploratory and confirmatory factor analyses, and reliability analyses, with multilevel data. The second assumption, that of structural invariance, is engineered into the common multilevel random coefficient model, in that such analyses impose structural invariance across multiple levels of analysis when lower-level relationships represent both between- and within-groups effects. The nature of these assumptions, and ways to address their tenability, are explored from a conceptual standpoint. Then an empirical example of these assumptions and ways to address them is provided. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The use of multiple imputation for the analysis of missing data.

This article provides a comprehensive review of multiple imputation (MI), a technique for analyzing data sets with missing values. Formally, MI is the process of replacing each missing data point with a set of m > 1 plausible values to generate m complete data sets. These complete data sets are then analyzed by standard statistical software, and the results combined, to give parameter estimates and standard errors that take into account the uncertainty due to the missing data values. This article introduces the idea behind MI, discusses the advantages of MI over existing techniques for addressing missing data, describes how to do MI for real problems, reviews the software available to implement MI, and discusses the results of a simulation study aimed at finding out how assumptions regarding the imputation model affect the parameter estimates provided by MI. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Speaking up in groups: A cross-level study of group voice climate and voice.

Despite a growing body of research on employee voice—defined as the discretionary communication of ideas, suggestions, or opinions intended to improve organizational or unit functioning—the effects of shared or collective-level cognitions have received scant attention. There has also been relatively little research on voice within work groups. Our goal in this study was to address these important gaps by focusing on the effects of group-level beliefs about voice (i.e., group voice climate) on individual voice behavior within work groups. We conducted a cross-level investigation of voice behavior within 42 groups of engineers from a large chemical company. Consistent with our hypotheses, group voice climate was highly predictive of voice and explained variance beyond the effects of individual-level identification and satisfaction, and procedural justice climate. Also consistent with predictions, the effect of identification on voice was stronger in groups with favorable voice climates. These findings provide evidence that voice is shaped not just by individual attitudes and perceptions of the work context, as past research has shown, but also by group-level beliefs. The results also highlight the importance of broadening our conceptual models of voice to include shared cognitions and of conducting additional cross-level research on voice. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The statistical concepts of confidence and significance.

"Inasmuch as explicit terminology is needed to convey the probabilities of committing statistical errors in the respective areas of interval estimation and testing hypotheses, the concept of confidence should never be associated with the statistical test of an H regardless of the nature of the test being employed." (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Utilities as base-rate multipliers in the determination of optimum cutting scores for the discrimination of groups of unequal size and variance.

The accuracy with which a test classified people, objects, or events as belonging to 1 of 2 groups depends upon the distance between the means, the relative variability, the relative size, and the shape of distributions of the 2 groups. If the scores for each of the groups are normally distributed, tables for determining optimum cutting scores for a wide range of values of the other variables are now available. However, overall accuracy is an appropriate guide for decision making only when all correct classifications are equally beneficial and all incorrect classifications equally costly. A simple technique makes possible the utilization of the Rorer, Hoffman, and Hsieh tables when a different value is assigned to each of the outcomes. (PsycINFO Database Record (c) 2010 APA, all rights reserved)