Dynamic aspects of voluntary turnover: An integrated approach to curvilinearity in the performance–turnover relationship.

Previous research pertaining to job performance and voluntary turnover has been guided by 2 distinct theoretical perspectives. First, the push–pull model proposes that there is a quadratic or curvilinear relationship existing between these 2 variables. Second, the unfolding model of turnover posits that turnover is a dynamic process and that a downward performance change may increase the likelihood of organizational separation. Drawing on decision theory, we propose and test an integrative framework. This approach incorporates both of these earlier models. Specifically, we argue that individuals are most likely to voluntarily exit when they are below-average performers who are also experiencing a downward performance change. Furthermore, the interaction between this downward change and performance partially accounts for the curvilinear relationship proposed by the push–pull model. Findings from a longitudinal field study supported this integrative theory. (PsycINFO Database Record (c) 2011 APA, all rights reserved)     

Bivariate median splits and spurious statistical significance.

Despite pleas from methodologists, researchers often continue to dichotomize continuous predictor variables. The primary argument against this practice has been that it underestimates the strength of relationships and reduces statistical power. Although this argument is correct for relationships involving a single predictor, a different problem can arise when multiple predictors are involved. Specifically, dichotomizing 2 continuous independent variables can lead to false statistical significance. As a result, the typical justification for using a median split as long as results continue to be statistically significant is invalid, because such results may in fact be spurious. Thus, researchers who dichotomize multiple continuous predictor variables not only may lose power to detect true predictor–criterion relationships in some situations but also may dramatically increase the probability of Type I errors in other situations. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Latent variable interaction and quadratic effect estimation: A two-step technique using structural equation analysis.

The author proposes an alternative estimation technique for latent variable interactions and quadratics. Available techniques for specifying these variables in structural equation models require adding variables or constraint equations that can produce specification tedium and errors or estimation difficulties. The proposed technique avoids these difficulties and may be useful for EQS, LISREL 7, and LISREL 8 users. First, measurement parameters for indicator loadings and errors of linear latent variables are estimated in a measurement model that excludes the interaction and quadratic variables. Next, these estimates are used to calculate values for the indicator loadings and error variances of the interaction and quadratic latent variables. Then, these calculated values are specified as constants in the structural model containing the interaction and quadratic variables. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Distinguishing between moderator and quadratic effects in multiple regression.

Moderated regression analysis is commonly used to test for multiplicative influences of independent variables in regression models. D. Lubinski and L. G. Humphreys (1990) have shown that significant moderator effects can exist even when stronger quadratic effects are present. They recommend comparing effect sizes associated with both effect types and selecting the model that yields the strongest effect. The authors show that this procedure of comparing effect sizes is biased in favor of the moderated model when multicollinearity is high because of the differential reliability of the quadratic and multiplicative terms in the regression models. Fortunately, levels of multicollinearity under which this bias is most problematic may be outside the range encountered in many empirical studies. The authors discuss causes and implications of this phenomenon as well as alternative procedures for evaluating structural relationships among variables. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Occupational status, family environments, and adolescents' aspirations: The Laosa model.

L. M. Laosa (see record 1983-11265-001) suggested a theoretical model in which parent–child interactions are a mediating variable between social-status indicators and children's attributes. The present study examined relationships between family learning environments and the aspirations of 512 Australian adolescents from 3 occupational status groups. Family environments were assessed initially when the adolescents were 11 yrs old, and measures were obtained from parents of their aspirations for their children and their instrumental and affective orientations toward learning. When the adolescents were 16 yrs old, their perceptions of the support for learning provided by their parents were measured. Regression surfaces were constructed from models that included terms to account for possible linear, interaction, and curvilinear associations among the variables. Adolescents' aspirations generally had moderate associations with parents' aspirations but only modest or negligible relations to parents' instrumental and affective orientations. Although occupational status had only modest or negligible associations with the environment and aspiration scores, results indicate that within the different occupational groups the environmental variables combined to have variable patterns of linear, interaction, and curvilinear relationships with adolescents' aspirations. (29 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Perceiver and relationship effects on perceptions of HIV status: A naturalistic study in a high drug use sample.

This article examines the effect of target, perceiver, and relationship characteristics on the perceiver's assessment that the target may be HIV seropositive (HIV+). A sample of 267 persons was recruited from low income, high drug use neighborhoods. Respondents (perceivers) were asked to name people (targets) with whom they had a social, drug sharing, or sexual relationship. Perceivers described 1,640 such relationships. Perceivers were asked about the targets' age, gender, and race/ethnicity, whether the targets were good-looking, their level of trust with the target, and how long they had known them. Perceivers were then asked to evaluate the chances that the target mentioned was HIV+. Two regression models were estimated on the 1,640 relationships mentioned. Model 1 included variables reflecting only target characteristics as independent variables. Model 2 included variables reflecting target characteristics as well as variables reflecting perceivers and perceiver-target relationship characteristics. The results showed that targets that were female, younger, and good-looking were perceived as being less likely to be HIV+. However, when accounting for perceiver and relationship effects, some of the target characteristic effects disappeared. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Molecular aspects of breast and ovarian cancer

Side lobes and grating lobes are both unwanted parts of the ultrasound beam emitted off axis that produce image artifacts due to error in positioning the returning echo. The purpose of this study was to reproduce artifacts associated with side lobes and grating lobes in vitro using different transducer types and recognize these artifacts in vivo. A phantom, composed of a water bath, a metallic wire, and a wooden tongue depressor, was imaged using a linear array, a curved linear a vector array, and a sector mechanical transducer. When imaging the metallic wire in a transverse plane, an echogenic artifact was constantly seen on each side of the wire, with a shape and intensity variable with the transducer type. The artifact was curvilinear and concave (linear and curved linear arrays), or curvilinear and convex (vector array and the mechanical transducer). When the tongue depressor was imaged in a longitudinal plane, the artifact was a straight line (linear array), a curved convex line (curved array), a series of convex curvilinear echo (vector array) or a small convex curvilinear echo (mechanical transducer). In vivo situations similar to the phantom experiment were investigated using clinical patients. Artifacts produced in vitro were recognized in vivo when a highly reflective object (urinary bladder wall) was imaged adjacent to an anechoic region (urine). These artifacts corresponded to the principle of secondary ultrasound lobes, and were therefore interpreted as such.     

On the Relationship Between Deformation-Induced Surface Roughness and Plastic Strain in AA5052—Is it Really Linear?

Three different heat treatments of aluminum alloy AA5052 were subjected to various levels of uniaxial plastic strain. The resulting surfaces were then evaluated using both scanning laser confocal microscopy (SLCM) and stylus profilometry. Three regression approaches were used to assess the quality of a linear and a curvilinear fit for the roughness data as a function of true plastic strain. While there were differences among the regression results, the analyses revealed that a linear model was more statistically appropriate for the finest grain size. As the grain size increased, the surface morphology became more complex and a quadratic model became more suitable. Since the relative area fractions of grain boundary–localized roughness and slip-induced roughness are grain size dependent, the higher order fit between the roughness and plastic strain reflects substantial changes in the ratio of these areas. The differences between the SLCM and profilometry results were attributed to the natural filtering that occurs during contact profilometry. This filtering skewed the roughness data toward the largest surface displacements, thereby reducing the measurement fidelity to the point where the only possible outcome was the linear relationship.     

Detection of convexity and concavity in context.

Sensitivity to shape changes was measured, in particular detection of convexity and concavity changes. The available data are contradictory. The author used a change detection task and simple polygons to systematically manipulate convexity/concavity. Performance was high for detecting a change of sign (a new concave vertex along a convex contour or a new convex vertex along a concave contour). Other things being equal, there was no evidence of an advantage for detecting a new concavity compared with a new convexity, for detecting a change of angle to a concave vertex compared with a convex vertex, for detecting a change within a concave region compared with a change within a convex region, or for an interaction between convexity and concavity and changes affecting or not affecting a vertex. The author concludes that change detection is affected by changes of sign of curvature (leading to changes in part structure). However, contrary to previous proposals, there is no special role for negative curvature or minima of curvature in guiding attention. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Sexualité, intimité et relation d'aide.

Examines sexual relationships in terms of their impact on the client, therapist variables, and the basic features of the helping relationship. Even though ethical formulations unanimously condemn sexual relationships between clients and therapists, it is suggested that the clients' interests require the articulation of an integrative and practical model of sexual interaction in therapeutic relationships. (English abstract) (34 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Regression models: calculating the confidence interval of effects in the presence of interactions

The main goal of regression analysis (multiple, logistic, Cox) is to assess the relationship of one or more exposure variables to a response variable, in the presence of confounding and interaction. The confidence interval for the regression coefficient of the exposure variable, obtained through the use of a computer statistical package, quantify these relationships for models without interaction. Relationships between variables that present interactions are represented by two or more terms, and the corresponding confidence intervals can be calculated 'manually' from the covariance matrix. This paper suggests an easy procedure for obtaining confidence intervals from any statistical package. This procedure is applicable for modifying variables which are continuous as well as categorical.     

Estimating the reliability of a test split into two parts of equal or unequal length.

When the reliability of test scores must be estimated by an internal consistency method, partition of the test into just 2 parts may be the only way to maintain content equivalence of the parts. If the parts are classically parallel, the Spearman-Brown formula may be validly used to estimate the reliability of total scores. If the parts differ in their standard deviations but are tau equivalent, Cronbach's alpha is appropriate. However, if the 2 parts are congeneric, that is, they are unequal in functional length or they comprise heterogeneous item types, a less well-known estimate, the Angoff-Feldt coefficient, is appropriate. Guidelines in terms of the ratio of standard deviations are proposed for choosing among Spearman-Brown, alpha, and Angoff-Feldt coefficients. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Covariates of infant mortality in China: an exploratory approach

To elucidate the nature of the relationship between infant mortality in China and a variety of covariates using data from the 2/1000 Chinese Fertility Survey, we use a logistic regression model where the covariates are transformed with the help of Alternating Conditional Expectation (ACE) algorithm. This approach is used to overcome the general problem in multivariate regression analysis of coding the independent variables so that relationship between independent variables and response variables is best described, rather than coding such variables in an arbitrary way. The study demonstrates the procedures and usefulness of the ACE guided transformation in multivariate analysis. The transformed covariates are then used to estimate the effects of a series of socioeconomic and demographic factors collected in the study of infant death in China. The study shows that after appropriate transformations, all the demographic and socioeconomic variables selected have statistically significant and direct influence on infant death.     

Relationship of preoperative fear, type of coping, and information recevied about surgery to recovery from surgery

This study investigated the relationship of preoperative level of fear, extent of information seeking (coping), and amount of information received about surgery to recovery from surgery. It was hypothesized that a curvilinear relationship would be found between level of preoperative fear and recovery and between extent of information seeking and recovery; it was also hypothesized that amount of preoperative information obtained would interact with extent of information seeking. The subjects were 57 female patients between the ages of 18 and 68 who were schedule for abdominal surgery. The recovery measures consisted of a self-rating of postoperative negative affect (fear, depression, and anger), number of postoperative analgesics and sedatives received, and total number of days to discharge. No curvilinear relationships were found between preoperative level of fear or type of coping and recovery from surgery. Results showed a linear relationship between level of preoperative fear and recovery, with the least favorable recovery associated with high levels of preoperative fear. A significant interaction was found between level of preoperative fear and amount of preoperative information, with high-fear subjects who reported little preoperative information experiencing the least favorable recovery period. The findings are discussed in terms of the parallel response model proposed by Leventhal.     

Stress at work: Linear and curvilinear effects of psychological-, job-, and organization-related factors: An exploratory study of trinidad and tobago.

In this study, the authors operationalized job stress as a two-dimensional construct consisting of time pressure and anxiety. The authors hypothesized that the relationship between job stress and job-related attitudes such as job involvement and job satisfaction would be curvilinear but would be linear with psychosomatic problems. In addition, the authors proposed that attitudinal factors would mediate the relationship between job stress and organizational commitment. Data were obtained from 241 respondents in Trinidad and Tobago. Our findings revealed that curvilinear relationships were supported for anxiety and the outcome variables but not for time pressure. The results also provided full support for our mediation hypotheses in the case of anxiety. However, partial support for mediation was obtained for time pressure. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Violence is a curvilinear function of temperature in Dallas: A replication.

Data on weather and aggravated assaults were obtained to determine whether the curvilinear relationship between temperature and violence previously observed in Minneapolis, Minnesota (E. G. Cohn & J. Rotton, 1997), could be replicated. The data consisted of calls for services received by police in Dallas between January 1, 1994, and December 31, 1995. Controlling for holidays, school closings, time of day, day of the week, season of the year, and their interactions, moderator-variable autoregression analyses indicated that assaults were an inverted U-shaped function of temperature. Replicating past research, the curvilinear relationship was dominant during daylight hours and spring months, whereas linear relationships were observed during nighttime hours and other seasons. The results are interpreted in terms of routine activity theory and the negative affect escape model of aggression. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Looking forward and looking back: Integrating completion and sunk-cost effects within an escalation-of-commitment progress decision.

Currently, there are 2 conflicting frameworks with which to understand why decision makers might escalate their commitment to a previously chosen course of action: sunk costs and project completion. The author proposes that sunk costs and need to complete exert simultaneous pressures, both independent and interactive, on a decision maker's level of commitment. The responses of 340 participants were analyzed and supported a complementary relationship between the 2 predictors. In addition, sunk costs demonstrated a curvilinear influence on commitment and an interaction with level of completion that supported a Level of Completion X Sunk Cost moderation model. (A marginal utility model was not supported.) Results are discussed in terms of their relevance toward offering a complementary view of 2 potential antecedents to a decision maker's propensity to escalate his or her commitment to a previously chosen course of action. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Ability and attitude correlates of academic achievement: Family-group differences.

In this study, I examined the proposition that for children from different family groups, there are variations in relations among their ability, attitudes toward school, and academic achievement. Data were collected from 928 11-year-old Australian children and their parents. In the analysis, the children were classified into four family groups that were defined conjointly by family social status and parents' getting-ahead or getting-by orientation. Within each family group, regression surfaces were constructed from models that included terms to account for possible linear, interaction, and curvilinear associations among the variables. The results indicated that (a) there are moderate family-group differences in children's word performance and more modest variations in their ability, attitudes toward school, and mathematics achievement and (b) ability and school attitudes had differential linear and curvilinear relations to academic achievement for boys and girls from different family groups. In general, the study indicated that the family may act as a critical substratum variable that influences the relations between children's attitudinal and cognitive attributes and their academic performance. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Regularities of source recognition: ROC analysis.

Source memory has become the focus of a growing number of investigations in a variety of fields. An appropriate model for source memory is, therefore, of increasing importance. A simple 2-dimensional signal-detection model of source recognition is presented. The receiver operating characteristics (ROCs) obtained from 3 experiments are then used to test the model. The data demonstrate 3 regularities: convex ROCs, z-ROCs with linear slopes of 1.00, and slightly concave z-ROCs. Two of these regularities support the model. The 3rd requires a revision of the model. This revised model is fitted to the data. The implications of these regularities for other theories are also discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Simple Curvilinear Method for Numerical Methods of Open Channels

Three-dimensional finite-difference or finite-volume models of sinuous open channels (e.g., narrow rivers, estuaries, and reservoirs) generally require boundary-fitted grids and curvilinear flow solution. Cartesian models with square grid cells are simpler to apply, but require a larger number of cells, as the cell size is determined by cross-stream resolution. This paper presents a simplified curvilinear approach suitable for systems where the along-stream length scale is larger than the cross-stream scale. The curvilinear Navier-Stokes equations are manipulated so the left-hand side is identical to the Cartesian momentum equations. The right-hand side then consists of grid-stretching curvature terms. These terms are written as functions of a perturbation parameter, so the first-order curvilinear effects are obtained with the lowest-order perturbation terms. As the Cartesian equations' form is preserved, we can readily adapt a Cartesian model to this perturbation curvilinear approach by adding the small curvilinear terms as explicit momentum sources.