An Empirical Study of Software Reuse with Special Attention to Ada

IEEE TRANSACTIONS on SOFTWARE ENGINEERING. VOL. 23. NO. 9. SEPTEMBER 1997

http://ieeexplore.ieee.org/iel3/32/13613/00629492.pdf?tp=&isnumber=&arnumber=629492

 

Abstract

The purpose of this empirical study is to examine how organizations actually exploit reuse technologies and evaluates how reuse factors affect the rate of reuse in an organization. This study is an attempt to enhance the measurement of the rate of reuse and the effectiveness of reuse by establishing conceptual foundations in the literature for reuse and conducting an empirical investigation of organizations using Ada technology. This study differentiated software reuse into six criteria: domain, human, tool, organization, software metrics, and environment. The results of this study show that the rate of reuse significantly depends upon reuse capability, software development effort, object-oriented design capability, repository development effort, Ada technology capability, and domain capability.

 

1. Introduction

- Reuse에 대한 배경설명

- 논문의 제약사항(limitation): Ada lang.에 한정, mail questionaire에 의한 결과수집(case study, field study에 비해서 상세하지 못함)

- 논문의 구성

 

2. Discussion of the origin of variables

- Durek's idea(unpublished paper)에 근간하여 6개의 category구성

(각각은 설문지의 질의사항로 연결됨)

- 각각의 category에 대한 기존의 대립되는 연구결과소개 및 본 논문에서 명료한 결과를 제시할 것을 약속

 

3. Research framework

3.1 Research questions

- 19개의 질의사항소개

- 7개의 reuse factor소개

(In order to answer these questions(설의응답결과를 활용하기위해서?), the authors developed a reuse model on the basis of the results of factor analysis on the collected data.(Multivariate Data Analysis with Readings, second edition, p. 189)

 

3.2 Research hypothesis

- 가설소개

- 가설의 전제사항(assumption)설명

- dependent variable설명(the rate of reuse)

 

3.3 Research Methodology

- random sampling technique사용(편차가 적고 일반적이라고 설명)

- mail survey사용(low-cost method)

 

4. Data analysis

4.1 Test of Nonresponse bias

- 무응답자가 있었으나 bias가 없었다는 것을 확인

- Two distributions of software size were employed for the test of nonresponse bias

: software size in an organization which the respondents were employed in

, software size of 500 organizations maintained by the Ada Usage Database of the AdaIC which was used as the population.

 

4.2 Test of reliability and validity

- Reliability can be defined broadly as the degree to which measures are free from error and, therefore, yield consistent results.

(설문대상조직의 OOD 및 Ada에 대한 경험치 및 전문성이 높기 때문에 문제가 없다고 설명(?))

- factor들간의 validity확인으로 제거할 것들이 없음을 확인

 

4.3 Demographic analysis

- 설문대상자 및 대상조직이 신뢰할 수 있는 안정된 대상임을 설명

 

4.4 Descriptive statistics

- 설문결과분석 및 계산(mean, Std Dev.)

 

4.5 Testing the hypothesis

- 7 factors analysis수행

The purpose of factor analysis is to analyze interrelationships among a large number of variables and to explain these variables in terms of their common underlying dimensions or factors. on the other hand, the objective or regression analysis is to help us predict a single dependent variable from the knowledge of one or more independent variables.

 

5. Findings and implications

- 6개 category 및 세부 항목들에 대해서 the rate of reuse에 미치는 영향정도 설명

(p-value: 0.1, 0.05)

 

6. Conclusions

- This study reports on systematic reuse used in organizations that are formally and consciously organized to deal with the software crisis.

- This study also confirms that reuse capability, OOD capability, repository, and domain capability are key success factors.