Four-wave Panel Research Articles

Working Long Hours and Career Wage Growth

This study establishes empirically a positive relationship between hours worked per week and growth in hourly wages. A four-wave panel survey of men and women who registered to take the Graduate Management Admission Test between June 1990 and March 1991 is used to show that this relationship is especially strong for young professional workers, but it is also present in a more wide-ranging dataset like the 1979 cohort of the NLSY. I nd the relationship to be nonlinear: in the GMAT Registrant Survey, for workers who put in 48 or more hours per week annual wage growth increases by 2 percent per 10 extra hours worked per week. The average eect is zero when hours are less than 47. The positive eect of hours on wage growth can be accounted for both by a learning-by-doing model with heterogeneous preferences for leisure, and by a model of promotions that combines higher per hour productivity in upper levels of the job ladder with worker heterogeneity. Using data on promotions and training, I provide evidence in support of the job-ladder model.

Assessing Causality in the Relationship Between Community Attachment and Local News Media Use

Numerous studies have demonstrated a relationship between community attachment and local news media use. Despite calls for panel studies to determine the direction of causality in this relationship, there is little evidence beyond cross-sectional surveys, which are often further limited to single communities. In order to contribute to the debate about causal direction, we conducted a four-wave national panel study with repeated measurement of community attachment and local news media use. Cross-sectional analyses confirmed the expected relationship between news use and community attachment. However, more conservative panel analyses controlling for a lagged measure of the dependent variable failed to produce evidence of causal relationships. The hypothesized role of population density and ethnic diversity as cross-level moderators of this relationship was examined using multilevel modeling. Methodological and theoretical reasons for the results are discussed and suggestions for alternative study designs are proposed.

The Dynamics of Political Protest: Feedback Effects and Interdependence in the Explanation of Protest Participation

This article addresses three largely unsolved problems in theory and research on political protest. The first problem concerns feedback effects. The common assumption is that protest is determined by various factors and does not influence its determinants. We propose and test hypotheses about feedback effects of protest on its determinants. The second issue is the usual assumption that the determinants do not influence each other. We propose and test hypotheses about their interdependence. The third issue which is also rarely addressed in the literature is explaining different effects of individual-level variables (i.e. coefficients) by changes of the political context. Since we test our hypotheses with a four-wave panel survey, conducted in Leipzig between 1990 and 1998, we suggest propositions about differential effects of the variables across waves. We find, among other things, that people having been engaged in protest activities under communist rule in 1989 tend to exhibit a long-term decrease of political discontent. Finally, we find that integration in protest-promoting networks is the most important determinant of protest.

Estimating regression to the mean and true effects of an intervention in a four‐wave panel study

First, to analyse whether a taxation-related decrease in spirit prices had a similar effect on spirit consumption for low-, medium- and high-level drinkers. Secondly, as the relationship between baseline values and post-intervention changes is confounded with regression to the mean (RTM) effects, to apply different approaches for estimating the RTM effect and true change. Consumption of spirits and total alcohol consumption were analysed in a four-wave panel study (one pre-intervention and three post-intervention measurements) of 889 alcohol consumers sampled from the general population of Switzerland. Two correlational methods, one method quantitatively estimating the RTM effect and one growth curve approach based on hierarchical linear models (HLM), were used to estimate RTM effects among low-, medium- and high-level drinkers. Adjusted for RTM effects, high-level drinkers increased consumption more than lighter drinkers in the short term, but this was not a persisting effect. Changes in taxation affected mainly light and moderate drinkers in the long term. All methods concurred that RTM effects were present to a considerable degree, and methods quantifying the RTM effect or adjusting for it yielded similar estimates. Intervention studies have to consider RTM effects both in the study design and in the evaluation methods. Observed changes can be adjusted for RTM effects and true change can be estimated. The recommended method, particularly if the aim is to estimate change not only for the sample as a whole, but for groups of drinkers with different baseline consumption levels, is growth curve modelling. If reliability of measurement instruments cannot be increased, the incorporation of more than one pre-intervention measurement point may be a valuable adjustment of the study design.

The Importance of Political Context for Understanding Civic Engagement: A Longitudinal Analysis

We contend that political context is important to consider when analyzing social capital and that context has an important but neglected impact on understanding the consequences of civic activity. Our focus is on the influence of rural, local leadership in two Minnesota communities and policies that these elites have developed to bring Internet connectivity to their citizens. One city developed a community electronic network and the other opted for an individualistic, entrepreneurial approach to information technology. Using a quasi-experimental research design and four-wave panel data, we find that elite policy approaches interact with civic activity to predict technology use among citizens, even long after the policies’ initial implementation. In the city with a community network, residents who are integrated into civic life are able to harness these political resources to become more technologically sophisticated.

Longitudinal Analysis of Adolescents’ Deviant and Delinquent Behavior

This article presents applications of different growth mixture models considering unobserved heterogeneity within the framework of Mplus ( Muthén & Muthén, 2001a , 2001b , 2004 ). Latent class growth mixture models are discussed under special consideration of count variables that can be incorporated into the mixtures via the Poisson and the zero-inflated Poisson model. Fourwave panel data from a German criminological youth study (Boers et al., 2002) is used for the model analyses. Three classes can be obtained from the data: Adolescents with almost no deviant and delinquent activities, a medium proportion of adolescents with a low increase of delinquency, and a small number with a larger growth starting on a higher level. Considering the zero inflation of the data results in better model fits compared to the Poisson model only. Linear growth specifications are almost sufficient. The conditional application of the mixture models includes gender and educational level of the schools as time-independent predictors that are able to explain a large proportion of the latent class distribution. The stepwise procedure from latent class growth analysis to growth mixture modeling is feasible for longitudinal analyses where individual growth trajectories are heterogenous even when the dependent variable under study cannot be treated as a continuous variable.

Are Charter School Parents More Satisfied With Schools? Evidence From Washington, DC

Parental satisfaction is critical to the politics of school reform. Many charter school advocates have no qualms in talking about a charter school movement, linking charter schools to a larger political effort to reform the system of education in the United States. If this movement is to have traction, parents who are satisfied with their children's schools represent a potential pool of supporters for further expansion of charter schools and choice in general. In contrast, if charter school parents are not satisfied with their schools, then a foundation for the movement is missing. In this article, using data from a four-wave panel study of parents in Washington, DC, with children in the traditional public schools and in the growing charter school sector, we explore the effect of charter school enrollment on satisfaction.

Development of deviant and delinquent behavior of adolescents

The article presents applications of different growth mixture models considering unobserved heterogeneity within the framework of Mplus (Muthén and Muthén, 2001, 2004). Latent class growth mixture models are discussed under special consideration of count variables which can be incorporated into the mixture models via the Poisson and the zero-inflated Poisson model. Four-wave panel data from a German criminological youth study (Boers et al., 2002) is used for the model analyses. Three classes can be obtained from the data: Adolescents with almost no deviant and delinquent activities, a medium proportion of adolescents with a low increase of delinquency and a small number with a larger growth starting on a higher level. The best model fits are obtained with the zero-inflated Poisson model. Linear growth specifications are almost sufficient. The conditional application of the mixture models includes gender and educational level of the schools as time-independent predictors which are able to explain a large proportion of the latent class distribution. The stepwise procedure from latent class growth analysis to growth mixture modeling is feasible for longitudinal analyses where individual growth trajectories are heterogenous even when the dependent variable under study cannot be treated as a continuous variable.

Television News, Mexico's 2000 Elections and Media Effects in Emerging Democracies

On the basis of an analysis of a four-wave panel survey, we argue that exposure to television news had significant, substantial effects on both attitudes and vote choices in Mexico's watershed presidential election of 2000. These findings support the contention, implicit in some research on political communication, that the magnitude of media effects varies with certain features of the political context. In particular, television influence in electoral campaigns may be substantially larger in emerging democratic systems.

How Does Postcommunist Transformation Affect Political Protest? The Example of East Germany

How do protest and its causes change after the democratic transformation of communist regimes? It is argued that certain features of the transformation process have an impact on the conditions that directly influence individual protest participation. These conditions are political discontent, perceived personal political influence, perceived obligation to participate, and integration in protest-.promoting networks. Drawing on the case of East Germany, propositions are suggested about how these four conditions change as a result of the transformation process, and about their differential impact on protest over time. The propositions are tested with a four-wave panel survey of 226 respondents from Leipzig (East Germany) who were interviewed in 1990 (referring to the situation under communist rule), 1993, 1996, and 1998. It was found, among other things, that protest decreased continuously between 1989 and 1998; and that each of the four factors' contribution to protest differed over time.

Trends of Criminal Activity and Substance Use in a Sample of Welfare Recipients

The Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) of 1996 instituted a compulsory work mandate for welfare recipients. However, recipients who experience difficulties finding employment may increase their involvement in criminal activities and their frequency of substance use as a means to deal with changes precipitated by PRWORA. This study used a four-wave panel design to analyze the criminal behaviors and substance-use frequency of 534 welfare recipients in Houston, Texas. Data were collected from the Attitudes, Behaviors, and Skills Assessment (ABSA) instrument designed specifically for this study. Results show that a minority of welfare recipients were involved in criminal activity. Furthermore, although participants were losing their welfare benefits, both criminal activity and substance use declined over time.

An Electorate Adrift?: Public Opinion and the Quality of Democracy in Mexico

When citizens lack stable political attitudes, leaders cannot easily be held accountable for their record in office, party system consolidation becomes more difficult, and public opinion is unable to offer much substantive guidance about policy-making. Ultimately, democratic governance is likely to suffer. In this article, we analyze a recent four-wave panel survey to assess the stability of political attitudes in Mexico. We find that the degree of attitude stability in Mexico varies across different types of dispositions. Although citizens hold reasonably firm views about the country's main political actors, preferences over issues are less consistent. These findings suggest both possibilities and constraints for demo- cratic governance.

Age and Distress in the Czech Republic

The relation between age and distress was examined with four-wave panel data from the Czech Republic. Age was positively associated with Czech men’s 1999 depression (the older panel members compared with the middle-aged members) and negatively with women’s problem behavior prior to controls. Age was also associated with change in men’s depression between 1994 and 1999, but not change in women’s problem behavior. The additions of life-course variables, economic experiences, and health and social-psychological resources helped explain the relation between age and change in men’s depression, and this was particularly true for economic experiences and personal resources. These same variables were also related to change in men’s problem behavior and women’s depression. Men’s problem behavior and women’s depression appeared to be distress symptoms but simply not related to age.

The Job Demands-Job Control Model and absence behaviour: Results of a 3-year longitudinal study

Empirical results of earlier studies only marginally supported the relevance of Karasek's Job Demands-Job Control Model for absence behaviour. Since longitudinal studies with respect to these relations were largely lacking, a four-wave panel study was carried out using data from 1755 male employees of a technical maintenance firm in the public sector. Job demands, job control, physical working conditions, and the employee's age, education, and health were measured in one year and absenteeism in the same year and in the next 3 years. Data were analysed with linear regression and Poisson regression techniques. The Poisson regression technique was superior to the linear regression technique in explaining absence. Age, health and prior absence were the best predictors of later absence behaviour. With respect to the Job Demands-Job Control Model, the main findings of the study were (1) that job control was significantly associated with a low number of simultaneous and later absence days, and (2) that, contrary to expectations, job demands were also related to a low number of simultaneous and later absence days. These results hold when age, health, education, prior absence, and working conditions are controlled for. Job control and job demands did not predict later absence frequency. In the discussion it is suggested that a high level of job demands may not only be harmful for the well-being of employees but also work as 'a pressure to attend'.

Mental Health and Desired Social Support: A Four-Wave Panel Study

The first aim of this study was to determine whether dissatisfaction with close (attachment) and less close (social integration) relationships was primarily reflected in desiring less or more social support. The second aim was to explore the temporal order of this association using latent variable LISREL analysis. Mental health was assessed by the-30-item General Health Questionnaire and desire for social support by the Interview Schedule for Social Interaction. These variables were measured at four consecutive waves each 4 months apart in a representative sample of 225 adult residents in Canberra. Psychological distress was associated with desiring more rather than less support. The LISREL analyses suggested that while mental health and desired level of attachment were unrelated, poorer previous mental health appeared to subsequently increase desired level of social integration.

An investigation of adolescents' substance use behaviors, depressed affect, and suicidal behaviors.

Data from a four-wave panel design of 975 adolescents were used to study inter-relationships among suicidal behaviors, depressive symptoms, and substance use behaviors. Persistently high levels of problem drinking and depressive symptoms were associated with higher levels of suicidal thoughts and attempts. Higher levels of depressive symptoms and greater cigarette and illicit drug use distinguished suicidal ideators from attempters. Adolescents attempting suicide reported lower levels of family social support, a greater use of substances to cope with stressors, and a higher density of substance-using peers. Implications of the findings for preventive interventions with high-risk teens are discussed.

Sources of CEO Power and Firm Financial Performance: A Longitudinal Assessment

The chief executive officer (CEO) is generally regarded as the most powerful organizational member. Attributions regarding the potential effect of a strong CEO on firm financial performance are common in both the academic literature and popular press. The oft-made assumption is that a powerful CEO will impact firm performance. The anticipated direction of this influence, however, has not been uniformly established. Moreover, we are not aware of any direct empirical assessment of the relationship between CEO power and firm financial performance. We are also intrigued by the potential for reciprocal effects—firm performance impacts the level of CEO’s power. Our review of the literature demonstrates no empirical attention to this possibility. This study relies on a four-wave panel design to assess the nature of the relationship between CEO power and firm performance. Results of LISREL analysis demonstrate that aspects of CEO power and financial performance are, in fact, interrelated. Performance was found to be both an antecedent condition and outcome of CEO power.

Sources of CEO power and firm financial performance: A longitudinal assessment

The chief executive officer (CEO) is generally regarded as the most powerful organizational member. Attributions regarding the potential effect of a strong CEO on firm financial performance are common in both the academic literature and popular press. The oft-made assumption is that a powerful CEO will impact firm performance. The anticipated direction of this influence, however, has not been uniformly established. Moreover, we are not aware of any direct empirical assessment of the relationship between CEO power and firm financial performance. We are also intrigued by the potential for reciprocal effects—firm performance impacts the level of CEO's power. Our review of the literature demonstrates no empirical attention to this possibility. This study relies on a four-wave panel design to assess the nature of the relationship between CEO power and firm performance. Results of LISREL analysis demonstrate that aspects of CEO power and financial performance are, in fact, interrelated. Performance was found to be both an antecedent condition and outcome of CEO power.

Alternative Methods for the Quantitative Analysis of Panel Data in Family Research: Pooled Time-Series Models

This article examines quantitative panel analysis techniques appropriate for situations in which the researcher models determinants of change in continuous outcomes. One set of techniques, based on the analysis of pooled time-series data sets, has received little attention in the family literature, although there are a number of situations where these techniques would be appropriate, such as the analysis of multiple-wave panel data. The article explores the fixed effect (change-scare) and random effects models based on pooled time-series data and discusses their advantages and disadvantages for the analysis of survey panel data compared with other available approaches. An empirical example of these methods is presented in which the effect of marital duration and number of children on spousal interaction in intact marriages was examined in a four-wave panel sample. A significant development in the field of family research has been the increased availability of large panel survey data sets containing variables of interest to family researchers (e.g., National Survey of Families and Households Panel, Panel Study of Income Dynamics, National Longitudinal Survey of Youth, Marital Instability Over the Life Course Panel). The potential posed by these panel studies is often unrealized because of the greater complexity of panel analyses and the shortage of clear and accessible guidelines for selecting the appropriate analysis models and statistical software. Several guides to panel methods are available (e.g., Campbell, Mutran, & Parker, 1986; Collins & Horn, 1991; Finkel, 1995; Johnson, 1988; Kessler & Greenberg, 1981; Markus, 1979; Menard, 1991). While these guides can be of value, they have the following drawbacks: Some recent developments are not covered, the focus has been largely on methods for two-wave panels, and situations commonly encountered by family researchers in the analysis of survey data, such as missing waves for some respondents, have been neglected. One type of model commonly found in family research involves the analysis of change over time in a continuous dependent variable. For example, a researcher interested in explaining change in marital interaction over the course of a marriage may model the effects of increased duration of the marriage, changes in the number and ages of the children, spells of employment of both spouses, and changes in family income. This situation involves a continuous dependent variable (degree of spousal interaction) with continuous variables (marital duration, income) and events (addition and subtraction of children, spells of employment and unemployment) as explanatory variables. Panel studies of changes in psychological distress brought about by marital dissolution, changes in frequency of sexual intercourse over the duration of the marriage, and the effect of retirement on marital happiness are research problems with similar analytic needs. With two or more waves of panel data and a continuous dependent variable, the researcher has a choice among five basic panel analysis models. These are (a) regression with lagged dependent variables, (b) structural equation models with reciprocal and lagged effects (e.g., LISREL), (c) repeated measures analysis of variance, (d) growth curve and hierarchical effects models, and (e) fixed (change-score) and random effects regression estimators for pooled time-series data sets. Event history, or hazard, models (Teachman, 1982) and panel models for qualitative variables (Clogg, Eliason, & Grego, 1990) are excluded from this list because they are not designed for use with continuous dependent variables. The first three techniques have been widely used with panel data in the family literature, and are generally accessible to the researcher. Many of the regression techniques have been superseded by the structural equation approaches, which are capable of estimating reciprocal effects and control for biases introduced by measurement errors and autocorrelated errors. …

The effect of characteristics of human service workers on subsequent morale and turnover.

The authors examine the relationship of sociopsychological characteristics of human services workers to their subsequent morale and continuance on the job. Eighty-two intensive case managers (half of them social workers) who had caseloads of clients with severe mental disorders were invited to participate in a four-wave panel study. Explored are the relationships between 16 predictors (general background, prior work experience and training, and social psychological variables) and worker outcomes (beliefs, attitudes, perceptions, and turnover). The major finding was that better personal well-being at the time of hiring was the most consistent and significant predictor of lower burnout, higher job satisfaction, and generally higher worker morale at later points in time. Those from higher social class backgrounds and those with lower starting salaries were more likely to leave the job. Those workers with more experience had a tendency to perceive greater client improvement.