Dimitra Dodou

Exploratory Factor Analysis With Small Sample Sizes

Exploratory factor analysis (EFA) is generally regarded as a technique for large sample sizes (N), with N = 50 as a reasonable absolute minimum. This study offers a comprehensive overview of the conditions in which EFA can yield good quality results for N below 50. Simulations were carried out to estimate the minimum required N for different levels of loadings (λ), number of factors (f), and number of variables (p) and to examine the extent to which a small N solution can sustain the presence of small distortions such as interfactor correlations, model error, secondary loadings, unequal loadings, and unequal p/f. Factor recovery was assessed in terms of pattern congruence coefficients, factor score correlations, Heywood cases, and the gap size between eigenvalues. A subsampling study was also conducted on a psychological dataset of individuals who filled in a Big Five Inventory via the Internet. Results showed that when data are well conditioned (i.e., high λ, low f, high p), EFA can yield reliable results for N well below 50, even in the presence of small distortions. Such conditions may be uncommon but should certainly not be ruled out in behavioral research data. * These authors contributed equally to this work

Factor recovery by principal axis factoring and maximum likelihood factor analysis as a function of factor pattern and sample size

Principal axis factoring (PAF) and maximum likelihood factor analysis (MLFA) are two of the most popular estimation methods in exploratory factor analysis. It is known that PAF is better able to recover weak factors and that the maximum likelihood estimator is asymptotically efficient. However, there is almost no evidence regarding which method should be preferred for different types of factor patterns and sample sizes. Simulations were conducted to investigate factor recovery by PAF and MLFA for distortions of ideal simple structure and sample sizes between 25 and 5000. Results showed that PAF is preferred for population solutions with few indicators per factor and for overextraction. MLFA outperformed PAF in cases of unequal loadings within factors and for underextraction. It was further shown that PAF and MLFA do not always converge with increasing sample size. The simulation findings were confirmed by an empirical study as well as by a classic plasmode, Thurstones box problem. The present results are of practical value for factor analysts.

Social desirability is the same in offline, online, and paper surveys

This meta-analysis compares social desirability between paper and computer surveys.Social desirability in computer (Internet/offline) and paper surveys is the same.Some of the large effects found in the past may have been due to sampling error. The aim of this meta-analysis was to compare social desirability scores between paper and computer surveys. Subgroup analyses were conducted with Internet connectivity, level of anonymity, individual or group test setting, possibility of skipping items, possibility of backtracking previous items, inclusion of questions of sensitive nature, and social desirability scale type as moderators. Subgroup analyses were also conducted for study characteristics, namely the randomisation of participants, sample type (students vs. other), and study design (between- vs. within-subjects). Social desirability scores between the two administration modes were compared for 51 studies that included 62 independent samples and 16,700 unique participants. The overall effect of administration mode was close to zero (Cohens d=0.00 for fixed-effect and d=-0.01 for random-effects meta-analysis). The majority of the effect sizes in the subgroup analyses were not significantly different from zero either. The effect sizes were close to zero for both Internet and offline surveys. In conclusion, the totality of evidence indicates that there is no difference in social desirability between paper-and-pencil surveys and computer surveys. Publication year and sample size were positively correlated (?=.64), which suggests that certain of the large effects that have been found in the past may have been due to sampling error.

Why the Fitts list has persisted throughout the history of function allocation

Function allocation is a core activity of the human–machine systems discipline. Sixty years ago, Paul Fitts marked the outset of function allocation research with an 11-statements list. Since then numerous function allocation methods have been proposed, but strikingly the seminal Fitts list spans the entire history of this domain and continues to be cited today. In this paper, we intend to explain why the Fitts list is such a pervasive factor in function allocation research, despite having received extensive criticism. We invoke philosophy of science, and we show that the Fitts list fulfils six important criteria for appraising scientific theories: plausibility, explanatory adequacy, interpretability, simplicity, descriptive adequacy, and generalisability. Furthermore, we show that the Fitts report identified issues which decades later became known as the ironies of automation. We conclude that the Fitts list is an adequate approximation that captures the most important regularity of automation, and that the Fitts report represents an unprecedented intellectual achievement that has succeeded in its pioneering objective.

Preparing drivers for dangerous situations: A critical reflection on continuous shared control

Shared control (also known as continuous haptic guidance or haptically active controls) has recently been introduced in car driving. With shared control, the driver receives continuous force feedback on the gas pedal or steering wheel, so that human and machine conduct the driving task simultaneously. Experiments in driving simulators have shown that shared control reduces control variability and mental workload, and improves accuracy in path tracking and car following. Crucial to road safety, however, is not whether shared control improves performance in routine driving tasks, but what happens in dangerous situations when a conflict of authority occurs, or when the force feedback cannot be relied upon or is suddenly disengaged. Drawing on research into transfer of training, it is shown that shared control may induce aftereffects and may hamper retention of robust driving skills. Supplementary information should not be provided continuously, but on an as-needed basis, warning or assisting drivers only when deviations from acceptable tolerance limits arise.

A surge of p-values between 0.041 and 0.049 in recent decades (but negative results are increasing rapidly too)

It is known that statistically significant (positive) results are more likely to be published than non-significant (negative) results. However, it has been unclear whether any increasing prevalence of positive results is stronger in the “softer” disciplines (social sciences) than in the “harder” disciplines (physical sciences), and whether the prevalence of negative results is decreasing over time. Using Scopus, we searched the abstracts of papers published between 1990 and 2013, and measured longitudinal trends of multiple expressions of positive versus negative results, including p-values between 0.041 and 0.049 versus p-values between 0.051 and 0.059, textual reporting of “significant difference” versus “no significant difference,” and the reporting of p < 0.05 versus p > 0.05. We found no support for a “hierarchy of sciences” with physical sciences at the top and social sciences at the bottom. However, we found large differences in reporting practices between disciplines, with p-values between 0.041 and 0.049 over 1990–2013 being 65.7 times more prevalent in the biological sciences than in the physical sciences. The p-values near the significance threshold of 0.05 on either side have both increased but with those p-values between 0.041 and 0.049 having increased to a greater extent (2013-to-1990 ratio of the percentage of papers = 10.3) than those between 0.051 and 0.059 (ratio = 3.6). Contradictorily, p < 0.05 has increased more slowly than p > 0.05 (ratios = 1.4 and 4.8, respectively), while the use of “significant difference” has shown only a modest increase compared to “no significant difference” (ratios = 1.5 and 1.1, respectively). We also compared reporting of significance in the United States, Asia, and Europe and found that the results are too inconsistent to draw conclusions on cross-cultural differences in significance reporting. We argue that the observed longitudinal trends are caused by negative factors, such as an increase of questionable research practices, but also by positive factors, such as an increase of quantitative research and structured reporting.

Common Factor Analysis versus Principal Component Analysis: A Comparison of Loadings by Means of Simulations

Common factor analysis (CFA) and principal component analysis (PCA) are widely used multivariate techniques. Using simulations, we compared CFA with PCA loadings for distortions of a perfect cluster configuration. Results showed that nonzero PCA loadings were higher and more stable than nonzero CFA loadings. Compared to CFA loadings, PCA loadings correlated weakly with the true factor loadings for underextraction, overextraction, and heterogeneous loadings within factors. The pattern of differences between CFA and PCA was consistent across sample sizes, levels of loadings, principal axis factoring versus maximum likelihood factor analysis, and blind versus target rotation.

Review of manual control methods for handheld maneuverable instruments.

Abstract By the introduction of new technologies, surgical procedures have been varying from free access in open surgery towards limited access in minimal access surgery. Improving access to difficult-to-reach anatomic sites, e.g. in neurosurgery or percutaneous interventions, needs advanced maneuverable instrumentation. Advances in maneuverable technology require the development of dedicated methods enabling surgeons to stay in direct, manual control of these complex instruments. This article gives an overview of the state-of-the-art in the development of manual control methods for handheld maneuverable instruments. It categorizes the manual control methods in three levels: a) number of steerable segments, b) number of Degrees Of Freedom (DOF), and c) coupling between control motion of the handle and steering motion of the tip. The literature research was completed by using Web of Science, Scopus and PubMed. The study shows that in controlling single steerable segments, direct as well as indirect control methods have been developed, whereas in controlling multiple steerable segments, a gradual shift can be noticed from parallel and serial control to integrated control. The development of multi-segmented maneuverable instruments is still at an early stage, and an intuitive and effective method to control them has to become a primary focus in the domain of minimal access surgery.

Buckling prevention strategies in nature as inspiration for improving percutaneous instruments: a review

A typical mechanical failure mode observed in slender percutaneous instruments, such as needles and guidewires, is buckling. Buckling is observed when the axial compressive force that is required to penetrate certain tissue types exceeds the critical load of the instrument and manifests itself by sudden lateral deflection of the instrument. In nature, several organisms are able to penetrate substrates without buckling while using apparatuses with diameters smaller than those of off-the-shelf available percutaneous needles and guidewires. In this study we reviewed the apparatuses and buckling prevention strategies employed by biological organisms to penetrate substrates such as wood and skin. A subdivision is made between buckling prevention strategies that focus on increasing the critical load of the penetration tool and strategies that focus on decreasing the penetration load of the substrate. In total, 28 buckling prevention strategies were identified and categorized. Most organisms appear to be using a combination of buckling prevention strategies simultaneously. Integration and combination of these biological buckling prevention strategies in percutaneous instruments may contribute to increasing the success rate of percutaneous interventions.

The Relationship Between Distal and Proximal Colonic Neoplasia: A Meta-Analysis

ABSTRACTOBJECTIVESTo investigate the association between proximal colonic neoplasia and distal lesions as a function of the lesion type. The extent to which health, demographic, and study characteristics moderate this association was also examined.DATA SOURCESGoogle Scholar, Web of Science, Scopus, and PubMed.STUDY ELIGIBILITY CRITERIAStudies allowing the calculation of OR of proximal neoplasia(PN) and proximal advanced neoplasia (PAN) for distal hyperplastic polyps (HP), nonadvanced adenomas (NAA), adenomas (AD), and advanced neoplasia (AN); also, studies for which the proportions of subjects with isolated (i.e., not accompanied by distal lesions) PN (IPN) and PAN (IPAN) over the total number of subjects with PN or PAN could be calculated.STUDY APPRAISAL AND SYNTHESIS METHODSThirty-two studies were included for calculating OR between proximal neoplasia and distal lesions and 40 studies for proportions of IPN and IPAN. Subgroup analyses were conducted for presence of symptoms, prevalence of PN and PAN, age, proportion of males, geographic region, study design, and demarcation point.RESULTSThe association between distal lesions and proximal neoplasia increased with the severity of the distal lesions. Odds of PN were higher in subjects with HP compared to subjects with a normal distal colon. Odds of PN and PAN were higher in subjects with NAA, AD, and AN than in subjects with a normal distal colon. PAN were more strongly associated with distal lesions in asymptomatic populations, in young populations, and in populations with a low prevalence of PAN. In approximately 60% of the subjects with PN and PAN, these neoplasia were isolated.LIMITATIONSThe present results may be affected by publication bias and dichotomization in the subgroup analyses. Limitations related to the individual studies include self-selection, lesion misclassification and misses, and technological advances leading to changes in the detection of lesions during the time span of the included studies.CONCLUSIONS AND IMPLICATIONS OF KEY FINDINGSAll types of distal lesions are predictive of PN. All types of distal neoplasia are predictive of PAN. The association between distal lesions and proximal neoplasia increases with the severity of the distal lesion. The association between distal lesions and proximal advanced neoplasia is stronger in low-risk groups as compared to high-risk groups.