東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

The career criminal debate: Comparin...

Burraston, Bert Owen.

FindBook

Google Book

Amazon

博客來

The career criminal debate: Comparing finite mixture modeling with growth mixture modeling.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	The career criminal debate: Comparing finite mixture modeling with growth mixture modeling./
作者:	Burraston, Bert Owen.
面頁冊數:	103 p.
附註:	Source: Dissertation Abstracts International, Volume: 64-06, Section: A, page: 2264.
Contained By:	Dissertation Abstracts International64-06A.
標題:	Sociology, Criminology and Penology. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3095238
ISBN:	0496428640

The career criminal debate: Comparing finite mixture modeling with growth mixture modeling.
Burraston, Bert Owen.

The career criminal debate: Comparing finite mixture modeling with growth mixture modeling. - 103 p.

Source: Dissertation Abstracts International, Volume: 64-06, Section: A, page: 2264.

Thesis (Ph.D.)--University of Oregon, 2003.

In this paper I compare recent methodological advances in the analysis of longitudinal data that have the potential to empirically test the Criminal Career Paradigm (CCP). I compare finite mixture modeling (FMM) with growth mixture modeling (GMM) and then apply both procedures to data from the Oregon Youth Study (OYS; Capaldi & Patterson, 1987), a longitudinal sample of 204 at-risk young men. Criminality was measured by the OYS boy's annual self-report of FBI index crimes committed from age 12 to 20.

ISBN: 0496428640Subjects--Topical Terms:

1017569
Sociology, Criminology and Penology.

The career criminal debate: Comparing finite mixture modeling with growth mixture modeling.
LDR:03473nmm 2200337 4500 001 1811991
005 20060329125111.5
008 130610s2003 eng d
020 $a 0496428640
035 $a (UnM)AAI3095238
035 $a AAI3095238
040 $a UnM $c UnM
100 1 $a Burraston, Bert Owen. $3 1901552
245 1 4 $a The career criminal debate: Comparing finite mixture modeling with growth mixture modeling.
300 $a 103 p.
500 $a Source: Dissertation Abstracts International, Volume: 64-06, Section: A, page: 2264.
500 $a Adviser: Robert O'Brien.
502 $a Thesis (Ph.D.)--University of Oregon, 2003.
520 $a In this paper I compare recent methodological advances in the analysis of longitudinal data that have the potential to empirically test the Criminal Career Paradigm (CCP). I compare finite mixture modeling (FMM) with growth mixture modeling (GMM) and then apply both procedures to data from the Oregon Youth Study (OYS; Capaldi & Patterson, 1987), a longitudinal sample of 204 at-risk young men. Criminality was measured by the OYS boy's annual self-report of FBI index crimes committed from age 12 to 20.
520 $a FMM has been used to test the CCP because of its ability to account for unobserved heterogeneity within a population and its ability to handle non-normally distributed data (Nagin and Land 1993). Jones, Nagin, and Roeder (2001) developed a SAS procedure (TRAJ) that runs FMM. Nagin and his colleagues have made substantial contributions to the CCP by providing a statistical modeling procedure that is capable of handling non-normal distributions and heterogeneity within a population. Their FMM, however, does not allow for within-class variation and therefore cannot test for heterogeneity.
520 $a Muthen (2001) has combined structural equation modeling with mixture modeling creating GMM. GMM can account for unobserved heterogeneity and can handle continuous, ordinal, categorical, and binary data. GMM allows for the simultaneous estimation of the antecedents of class membership.
520 $a Both methods support the CCP, however, the GMM procedure resulted in fewer classes. The FMM procedure resulted in a five class solution: (1) zero intercept non-significant slope, (2) near zero intercept non-significant slope, (3) near zero intercept with significant linear growth, (4) high intercept with a positive slope and negative quadratic slope, and (5) high intercept with a steep slope and a negative quadratic slope. The groups were distinguished by differences in parental discipline, monitoring, deviant peer association, family income, socioeconomic status, family structure, and sibling conflict.
520 $a The GMM procedure resulted in a three class solution: (1) zero intercept with zero slope, (2) low intercept with non-significant slope, and (3) high intercept with a steep positive slope and negative quadratic slope. The first 2 groups were distinguished by parental discipline, family income, and family structure while the final group was predicted by deviant peer association. I discuss the advantages and disadvantages of GMM and FMM procedures in the context of the Career Criminal Paradigm.
590 $a School code: 0171.
650 4 $a Sociology, Criminology and Penology. $3 1017569
650 4 $a Sociology, Theory and Methods. $3 626625
650 4 $a Sociology, Individual and Family Studies. $3 626655
690 $a 0627
690 $a 0344
690 $a 0628
710 2 0 $a University of Oregon. $3 958250
773 0 $t Dissertation Abstracts International $g 64-06A.
790 1 0 $a O'Brien, Robert, $e advisor
790 $a 0171
791 $a Ph.D.
792 $a 2003
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3095238