Development and Testing of an Instrument for Summative Assessment of Practical Skill Performance: A Generalizability Theory Approach

被引：2

作者：

Bjork, Ida Torunn ^{[1
]}

Larsen, Karin ^{[2
]}

Ravik, Monika ^{[3
]}

Reierson, Inger Ase ^{[3
]}

Sommer, Irene ^{[2
]}

Stenholt, Britta ^{[4
]}

Brynildsen, Grethe ^{[5
]}

Blomberg, Karin ^{[6
]}

Bolenius, Karin ^{[7
]}

Husebo, Sissel Eikeland ^{[8
,9
]}

Lamas, Kristina ^{[7
]}

Hagtvet, Knut Arne ^{[10
]}

机构：

[1] Univ Oslo, Oslo, Norway

[2] Aarhus Univ Hosp, Aarhus N, Denmark

[3] Univ South Eastern Norway, Porsgrunn, Norway

[4] VIA Univ Coll, Nurse Educ, Aarhus N, Denmark

[5] Lovisenberg Diaconal Univ Coll, Oslo, Norway

[6] Orebro Univ, Orebro, Sweden

[7] Umea Univ, Umea, Sweden

[8] Univ Stavanger, Stavanger, Norway

[9] Stavanger Univ Hosp, Stavanger, Norway

[10] Univ Oslo, Dept Psychol, Oslo, Norway

来源：

JOURNAL OF NURSING MEASUREMENT | 2021年 / 29卷 / 03期

关键词：

practical skill performance; multifacet summative assessment; generalizability estimation; nursing education; NURSING-STUDENTS; NURSES; COMPETENCE; EDUCATION; MODEL; INTERVENTION; PERCEPTIONS; RELIABILITY; EXPERIENCES; VALIDATION;

D O I：

10.1891/JNM-D-20-00003

中图分类号：

R47 [护理学];

学科分类号：

1011 ;

摘要：

Background and Purpose: Many newly graduated nurses lack proficiency in practical skill performance. Presently, nursing students' practical skill is assessed by summative instruments with overarching items. The purpose of this study was to develop a more detailed instrument to use in summative assessment of nursing students' practical skill performance and to assess its psychometric properties. Methods: A 50-item instrument was developed. Video-recorded performances were rated by experienced clinical supervisors. A multifacet measurement design was developed. Relevant parameters were estimated by generalizability analysis. Results: Findings indicated that error of measurement were mainly caused by raters far more than by items. Conclusions: The present study suggested that summative assessment in realistic settings may not apply one rater only. Two to three/four raters appear necessary to dependably measure most skills.

引用

页码：E162 / E191

页数：30

共 43 条

[21]

Crick J. E., 1983, AM COLL TESTING TECH, V43

[22]

CRONBACH L J, 1972, P410

[23] THEORY OF GENERALIZABILITY - A LIBERALIZATION OF RELIABILITY THEORY [J].