國立虎尾科技大學 |

A Monte Carlo simulation to examine the effects of violating the normality assumption in 2-level hierarchical linear models with unbalanced designs.

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	A Monte Carlo simulation to examine the effects of violating the normality assumption in 2-level hierarchical linear models with unbalanced designs./
作者:	Ketelsen, Kirk L.
面頁冊數:	1 online resource (144 pages)
附註:	Source: Dissertations Abstracts International, Volume: 76-05, Section: B.
Contained By:	Dissertations Abstracts International76-05B.
標題:	Statistics. -
電子資源:	click for full text (PQDT)
ISBN:	9781321321951

A Monte Carlo simulation to examine the effects of violating the normality assumption in 2-level hierarchical linear models with unbalanced designs.
Ketelsen, Kirk L.

A Monte Carlo simulation to examine the effects of violating the normality assumption in 2-level hierarchical linear models with unbalanced designs. - 1 online resource (144 pages)

Source: Dissertations Abstracts International, Volume: 76-05, Section: B.

Thesis (Ph.D.)--University of Northern Colorado, 2014.

Includes bibliographical references

Hierarchical Linear Modeling (HLM) is emerging as a preferred method for analyzing nested data when a hierarchical data structure exists. The purpose of this dissertation was to examine the performance of HLM under varying conditions commonly found in educational research. A Monte Carlo simulation was conducted to examine how varying levels of level-1 imbalance, intraclass correlation coefficients level-2 sample size, and non-normal level-2 error distribution would affect the fixed parameter estimates and their standard errors, type I error rates, and power. The HLM performed best when level-2 sample size was large, ICC was small, and level-2 error distribution was normal. Overall, fixed parameter estimates remained accurate under all conditions, however, standard errors were over-inflated when the level-2 error distribution was highly skewed (chi-squared). Power was acceptable for most design conditions except when the ICC was very large (.25). Implications for applied researchers include how to optimally design their studies (large level-2 sample size) to have sufficient power, as well as for checking for potential type II errors when level-2 error distribution is skewed.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9781321321951Subjects--Topical Terms:

556824
Statistics.
Subjects--Index Terms:

Hierarchical linear modelingIndex Terms--Genre/Form:

554714
Electronic books.

A Monte Carlo simulation to examine the effects of violating the normality assumption in 2-level hierarchical linear models with unbalanced designs.
LDR:02761ntm a22004097 4500 001 1152973
005 20241216064811.5
006 m o d
007 cr mn ---uuuuu
008 250605s2014 xx obm 000 0 eng d
020 $a 9781321321951
035 $a (MiAaPQ)AAI3644138
035 $a (MiAaPQ)unco:10350
035 $a AAI3644138
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Ketelsen, Kirk L. $3 1480099
245 1 2 $a A Monte Carlo simulation to examine the effects of violating the normality assumption in 2-level hierarchical linear models with unbalanced designs.
264 0 $c 2014
300 $a 1 online resource (144 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 76-05, Section: B.
500 $a Publisher info.: Dissertation/Thesis.
500 $a Advisor: Hutchinson, Susan R.
502 $a Thesis (Ph.D.)--University of Northern Colorado, 2014.
504 $a Includes bibliographical references
520 $a Hierarchical Linear Modeling (HLM) is emerging as a preferred method for analyzing nested data when a hierarchical data structure exists. The purpose of this dissertation was to examine the performance of HLM under varying conditions commonly found in educational research. A Monte Carlo simulation was conducted to examine how varying levels of level-1 imbalance, intraclass correlation coefficients level-2 sample size, and non-normal level-2 error distribution would affect the fixed parameter estimates and their standard errors, type I error rates, and power. The HLM performed best when level-2 sample size was large, ICC was small, and level-2 error distribution was normal. Overall, fixed parameter estimates remained accurate under all conditions, however, standard errors were over-inflated when the level-2 error distribution was highly skewed (chi-squared). Power was acceptable for most design conditions except when the ICC was very large (.25). Implications for applied researchers include how to optimally design their studies (large level-2 sample size) to have sufficient power, as well as for checking for potential type II errors when level-2 error distribution is skewed.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Statistics. $3 556824
653 $a Hierarchical linear modeling
653 $a Level-2 error distribution
653 $a Monte carlo simulation
653 $a Robustness
653 $a Unbalanced designs
653 $a Violation of assumptions
655 7 $a Electronic books. $2 local $3 554714
690 $a 0463
710 2 $a University of Northern Colorado. $b Applied Statistics & Research Methods. $3 1480100
710 2 $a ProQuest Information and Learning Co. $3 1178819
773 0 $t Dissertations Abstracts International $g 76-05B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3644138 $z click for full text (PQDT)