دانلود مقاله ISI انگلیسی شماره 34176
عنوان فارسی مقاله

تجزیه و تحلیل رگرسیون کارت با واحد وزنی برای پیش بینی افکار خودکشی از صفات پنج عامل بزرگ

کد مقاله سال انتشار مقاله انگلیسی ترجمه فارسی تعداد کلمات
34176 2003 13 صفحه PDF سفارش دهید محاسبه نشده
خرید مقاله
پس از پرداخت، فوراً می توانید مقاله را دانلود فرمایید.
عنوان انگلیسی
CART analysis with unit-weighted regression to predict suicidal ideation from Big Five traits
منبع

Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)

Journal : Personality and Individual Differences, Volume 35, Issue 2, July 2003, Pages 249–261

کلمات کلیدی
شخصیتی - مدل پنج عاملی - افکار خودکشی - واحد وزنی رگرسیون - طبقه بندی و تجزیه و تحلیل رگرسیون درختی -
پیش نمایش مقاله
پیش نمایش مقاله تجزیه و تحلیل رگرسیون کارت با واحد وزنی برای پیش بینی افکار خودکشی از صفات پنج عامل بزرگ

چکیده انگلیسی

This study examines the Big Five traits as multiple predictors of suicidal ideation. In addition to using multiple regression, this study introduces the use of classification and regression tree (CART) analysis to identify cutoff scores that can be used with unit-weighted regression; the purpose of this approach is to offer a method of analysis that may be useful in a public health approach to suicide prevention. The models were developed with a derivation group (N=299) and applied to a validation group (N=175). Both models performed well in the validation group, with multiple regression correlating at 0.47, and unit-weighted regression at 0.48. The models agreed in suggesting that increased suicidal ideation is associated with high Neuroticism, low Extraversion, and low Agreeableness. Low Conscientiousness was correlated with suicidal ideation in unit-weighted regression.

مقدمه انگلیسی

Suicide is a significant mental health problem. According to the National Center for Health Statistics (2001), suicide was the number four cause of years of life lost in the United States in 1998 (following cancer, heart disease, and accidental injuries), with an age-adjusted 363 years of potential life lost per 100,000 people. Personality traits may play an important role in understanding suicidal behavior. The traits of extraversion and neuroticism have been often studied. Higher levels of neuroticism have been found to correlate with more suicide attempts (Beautrais et al., 1999, Nordstrom et al., 1995, O’Boyle & Brandon, 1998 and Pallis & Jenkins, 1977), with more suicidal ideation (Lolas et al., 1991 and Velting, 1999), or with both ideation and attempts (Fergusson et al., 2000 and Statham et al., 1998). Low extraversion has been correlated with suicidal ideation (Lolas et al., 1991) and with a history of suicide attempts (Beautrais et al., 1999 and Roy, 1998) Several studies have also found that the trait of psychoticism from Eysenck’s personality system is significantly related to suicide (Ashton et al., 1994, Csorba et al., 1994, Engstrom et al., 1996, Lolas et al., 1991, Nordstrom et al., 1995 and Upmanyu et al., 1995). On a related note, Knight, Furnham, and Lester (2000) found that higher psychoticism scores correlated with more positive attitudes toward suicide. Given that Eysenck (1992) has argued that Psychoticism in his system is related to Agreeableness and Conscientiousness in the Big Five system, one might expect on theoretical grounds that suicidal ideation would be related to either low Agreeableness or low Conscientiousness, or perhaps both. However, despite this theoretical expectation, few studies have addressed the relationship of low Agreeableness or low Conscientiousness to suicidal ideation. One study of older adults (Duberstein, Conwell, & Caine, 1994) found no difference between suicide completers and controls on Conscientiousness or Agreeableness. But Velting (1999) found in a sample of undergraduates that low Conscientiousness was associated with higher suicidal ideation. Clearly, this issue needs more study. Thus, one purpose of this paper is to examine the Big Five traits as predictors of suicidal ideation, with specific attention to the hypothesis that low Agreeableness and low Conscientiousness would correlate with higher levels of suicidal ideation. A second purpose of this paper is to propose a data analysis method that may be useful in an applied setting where the goal is prevention. The prevention of mental health problems, as opposed to treatment after the problems occur, is the approach taken in public health (Galavotti et al., 1997, Miller et al., 1982, Roberts et al., 1997, Singer & Krantz, 1982 and Tanabe, 1982). And a public health approach to suicide in particular has been gaining attention recently (Durenberger, 1989, Mercy & Rosenberg, 2000, Potter et al., 1995, Potter et al., 1998 and Rutz, 2001). To apply a public health approach to suicide, one begins by identifying significant predictors (Centers for Disease Control, 1992). One common method for combining multiple predictors to predict an outcome is multiple regression. While it is a useful method, multiple regression has limitations for an applied prevention program. One limitation is that it assumes model specification; that is, it assumes that all the significant variables are in the model. This assumption is unlikely to be met in applied settings. Another limitation of multiple regression is that it offers no guidance on cutoff scores. For example, suppose that school officials plan a program to address suicidal thinking. Because previous research shows that higher neuroticism relates to more suicidal thinking, the officials plan a program to assist students in emotional coping. The applied question then becomes: what neuroticism score would suggest that a student might be at increased risk for suicidal thinking and might benefit from the emotional coping program? Multiple regression fails to answer this question; it identifies significant predictors, but not cutoff scores.So the second purpose of this study is to suggest a data analysis method that would address these two limitations of multiple regression; that is, the lack of cutoff scores, and the assumption of model specification. The use of a second method of analysis would allow one to follow the advice of Tukey (1969), who wrote, “One body of data can—and usually should—be analyzed in more than one way” (p. 83). The proposed second method for the current study has two steps. Step one provides cutoff scores by using classification and regression tree (CART) analysis (Breiman, Friedman, Olshen, & Stone, 1984). Step two avoids the assumption of model specification by combining the CART scores with unit-weighted regression (Dawes, 1979). Thus, this second method of analysis can be described as unit-weighted regression with CART scores. In summary, this study has two goals. First, the relationship between the Big Five traits and suicidal ideation is examined. Based on previous studies, it is hypothesized that high Neuroticism and low Extraversion will be associated with higher suicidal ideation. Based on the reported relationship between Psychoticism and suicidal ideation, it is also hypothesized that low Agreeableness and low Conscientiousness will be related to increased suicidal ideation. The second goal of the study is to assess the usefulness of unit-weighted regression with CART scores as a method of using personality traits as multiple predictors. Unit-weighted regression with CART scores does not assume model specification, so it is parsimonious; also, the method may be helpful in suggesting decisions in applied settings because of its identification of cutoff scores.

خرید مقاله
پس از پرداخت، فوراً می توانید مقاله را دانلود فرمایید.