A feasible method for standard errors of estimate in maximum likelihood factor analysis

A jackknife-like procedure is developed for producing standard errors of estimate in maximum likelihood factor analysis. Unlike earlier methods based on information theory, the procedure developed is computationally feasible on larger problems. Unlike earlier methods based on the jackknife, the present procedure is not plagued by the factor alignment problem, the Heywood case problem, or the necessity to jackknife by groups. Standard errors may be produced for rotated and unrotated loading estimates using either orthogonal or oblique rotation as well as for estimates of unique factor variances and common factor correlations. The total cost for larger problems is a small multiple of the square of the number of variables times the number of observations used in the analysis. Examples are given to demonstrate the feasibility of the method.The research done by R. I. Jennrich was supported in part by NSF Grant MCS 77-02121. The research done by D. B. Clarkson was supported in part by NSERC Grant A3109.     

Phenomenological research on supervision: Supervisors reflect on ‘being a supervisor’

A study using a phenomenological research method was performed to reveal the meanings of the concept ‘supervision’, from the supervisor's perspective. Eleven supervisors of counselling and psychotherapy from a humanistic/existential framework answered the open question: “What does being a supervisor mean?”. Content analysis and frequency measuring were performed, revealing six groups of statements; six facets describing the experience of supervision: ‘structuring’, ‘teaching’, ‘nurturing’, the ‘supervisor as person’, ‘supervisor as colleague’, and the ‘triangle, client-therapist-supervisor’. Structuring was the largest category, and the first three categories together represented more than 75% of the statements. The results showed that ‘teaching’ and ‘nurturing’ were very similar in their importance, as represented by the number of statements in each. Further research should address the possible relation between supervisors' perception and experience of their role and the theoretical framework of counselling and psychotherapy in which they are based, The phenomenological research method in this study has shown the possibility of doing research on processes without ‘betraying’ the subjective perspective.     

Working with countertransference

The transference-countertransference relationship is only one of five modalities of relationship that research has identified as potentially present in the therapeutic encounter. This paper gives the background and definition to one aspect of this - the countertransference - and traces the development of the concept from Freud's first use of the term in 1910 to the contemporary view that it is a useful tool of psychotherapy. The first part explains its connection with the Kleinian concept of projective identification and discusses its elaboration by the object relations school. There is general acceptance nowadays that the countertransference contains a great deal of information about the client's psychological world. It is therefore important to understand this process and the authors have identified three main dimensions to countertransference. These are its vector (or direction and force), its variance (the quality it represents), and its valence (its effect on the client). The second part of the article illustrates, through the use of example and metaphor, how these three dimensions are defined and can be recognized. Common themes and paradigms of countertransference are identified and discussed along with some ways in which experience has shown how these might be contained and worked with constructively. Finally, a clinical vignette is presented in which some of the dimensions of countertransference are identified and used to understand the client's psychic world and foster therapeutic change.     

The malleable influence of social consensus on attitude certainty

People often reflect on the opinions of others and express greater attitude certainty when they perceive their attitudes to be shared by others (high attitude consensus). The present research tests the possibility that either high or low attitude consensus can increase attitude certainty depending on people's salient social identification needs. In particular, high attitude consensus with a target group is found to be more validating when people seek to belong to the group, as this identification motive promotes a search for similarities between themselves and the group. In contrast, low attitude consensus with a target group is found to be more validating when people seek to be unique from a group, as this identification motive promotes a search for dissimilarities between themselves and the group. Two experiments support these hypotheses, offering insight into the intra-personal motives that alter the diagnostic value of social consensus information.     

Applying multiple models to predict clinicians’ behavioural intention and objective behaviour when managing children's teeth

This study used multiple theoretical approaches simultaneously to predict an objectively measured clinical behaviour. The six theoretical approaches were: The Theory of Planned Behaviour (TPB), Social Cognitive Theory (SCT), Common Sense Self-Regulation Model (CS-SRM), Operant Learning Theory (OLT), Action Planning (AP) and the Precaution Adoption Process (PAP), with knowledge as an additional predictor. Data on variables from these models were collected by postal survey. Data on the outcome behaviour, the evidence-based practice of placing fissure sealants, was collected from clinical records. Participants were 133 dentists (64% male) in Scotland. Variables found to predict the behaviour were: intention, attitude, perceived behavioural control, risk perception, outcome expectancies, self efficacy, habit, anticipated consequences, experienced consequences and action planning. The TPB, SCT, AP, OLT and PAP significantly predicted behaviour but the CS-SRM did not. A combined (Stepwise) regression model included only intention and action planning. Post hoc analyses showed action planning mediated effect of intention on behaviour. Taking a theory-based approach creates a replicable methodology for identifying factors predictive of clinical behaviour and for the design and choice of interventions to modify practice as new evidence emerges, increasing current options for improving health outcomes through influencing the implementation of best practice.     

Empirical support for the gender‐as‐culture hypothesis.

This investigation provided a test of the gender‐as‐culture, or ‘two cultures’, hypothesis proposed by Maltz and Borker (1982) to explain male/female differences in language use. Analysis of previous empirical investigations located 16 language features that had consistently been shown to indicate communicator gender and these were tested within the framework of the four dimensions of intercultural style proposed by Gudykunst and Ting‐Toomey (1998): direct versus indirect, succinct versus elaborate, personal versus contextual and instrumental versus affective. Study 1 provided preliminary evidence supporting the hypothesized language‐feature‐by‐dimension relationships (e.g., male directives were rated more direct and female uncertainty verbs more indirect). In Study 2, respondents rated multiple exemplars of the 16 language features, as well as 16 contrasting foil sentences, on all four dimensions, finding that nearly all of the variables fell on the hypothesized intercultural dimensions. In Study 3, respondents rated four sets of naturally occurring target sentences and matching foil sentences, representing all language variables, on their appropriate intercultural dimensions in order to establish dimensional polarity. Results across the three studies supported the hypothesized language feature‐by‐stylistic dimension relationship for 15 of the 16 variables: The 6 male language features were rated as more direct, succinct, personal, and instrumental, whereas 9 of the 10 female features were perceived as more indirect, elaborate, and affective. The findings demonstrate that gender preferences for language use function in ways that are consistent with stylistic preferences that distinguish national cultures.     

Theory and Practice: Narratives of psychotherapy in modern English literature

Narrative offers us a unique opportunity to enter an imaginative space in which to make sense of experiences and emotions in a playful way. In this study we posed the question: ‘How do novelists portray counselling in contemporary fiction?’. We found that the most effective practitioners were those who formed a strong, meaningful therapeutic relationship with their clients and who adopted a reflective, critical stance on their own practice. With regard to theoretical orientation, the most favourable accounts were of existentialist–phenomenological and cognitive–behavioural approaches. Client-centred and psychodynamic approaches were mainly portrayed negatively. Overall the importance of the therapeutic relationship as overriding considerations of schoolism was confirmed. The implications for psychotherapy and psychological counselling are discussed.     

The phonological neighbourhood effect on short-term memory for order

There is a growing body of literature that suggests that long-term memory (LTM) and short-term memory (STM) structures that were once thought to be distinct are actually co-dependent, and that LTM can aid retrieval from STM. The mechanism behind this effect is commonly argued to act on item memory but not on order memory. The aim of the current study was to examine whether LTM could exert an influence on STM for order by examining an effect attributed to LTM, the phonological neighbourhood effect, in a task that reduced the requirement to retain item information. In Experiment 1, 18 participants completed a serial reconstruction task where neighbourhood density alternated within the lists. In Experiment 2, 22 participants completed a serial reconstruction task using pure lists of dense and sparse neighbourhood words. In Experiment 3, 22 participants completed a reconstruction task with both mixed and pure lists. There was a significant effect of neighbourhood density with better recall for dense than sparse neighbourhood words in pure lists but not in mixed lists. Results suggest that LTM exerts an influence prior to that proposed by many models of memory for order.     

Two-year stability of personality disorder in older adolescent outpatients

The 2-year stability of categorical and dimensional personality disorder (PD) in an older adolescent psychiatric outpatient sample was examined. One hundred and one 15-18-year-old participants were assessed using the Structured Clinical Interview for DSM Axis II Disorders (SCID-II) at baseline and 97 were re-interviewed, face-to-face, at 2 years. Of those with a categorical PD diagnosis at baseline, 74% still met criteria for a PD at follow-up, with marked gender differences (83% of females and 56% of males). Kappa for specific PDs was low for all except antisocial. Rank order and mean level dimensional stability ranged from high (antisocial, schizoid) to moderate (borderline, histrionic, schizotypal) to low (other PDs), with no decline in PD scores over the 2 years. There was no substantial influence upon stability of dimensional PD from the presence of Axis I disorder at baseline or from outpatient or inpatient treatment. However, categorical PD endured in 100% of those receiving inpatient care. The study supports that, in late teenage outpatients, the 2-year stability of the global category of PD is high and the stability of dimensionally rated PD appears to be similar to that found in young adults in a variety of settings, especially for some cluster A and B PDs. Diagnosis and early intervention appears to be justified in this age group.     

Anticipatory Processing and Post‐Event Processing in Social Anxiety Disorder: An Update on the Literature

Two repetitive thinking processes that have been proposed in prominent maintenance models of social anxiety disorder (SAD) are anticipatory processing and post‐event processing. Research into these two processes has steadily increased over the last 20 years. This review highlights the main lines of existing research on anticipatory processing and post‐event processing, including studies on the nature of these processes, their association with social anxiety, the predictors, and consequences of these processes, as well as how these processes respond to treatments for SAD. The review also highlights some of the conceptual and methodological issues that have prevented the literature on anticipatory processing and post‐event processing from being more integrated and focused. Finally, the review draws together some new directions in terms of theory and research to further advance the field.