Earticle

집단 비교 연구 시 측정동일성의 성립 여부는 집단 간 의미 있는 비교를 하기 위한 필수 요건으로 제시되고 있다. 이를 위해 일반적으로 다집단 확인적요인분석(MG CFA)이 널리 사용되어 왔으나, MG CFA는 비교집단이 소수일 경우에 적합한 것으로, 많은 집단을 비교하게 되는 국가 비교 연구에서는 그 한계가 제기된다. 따라서 본 연구에서는 10개 이상의 많은 수의 국가(혹은 집단)를 비교하기에 적절한 방법론인 다층 확인적요인분석(ML CFA)과 다층 요인혼합모형(ML FMM)을 이용한 분석 방법론을 기술하였다. ML CFA는 절편만을 임의효과로 추정하는 임의절편모형과 요인계수도 임의효과로 추정하는 임의절편 및 임의요인계수모형으로 구분하여 장단점을 기술하였다. 구체적으로 각 방법론에 대한 이론적 모형과 측정동일성 검정 절차를 제시하고, 기존의 MG CFA에 비해 지니는 이점 및 적용 시 유의해야 할 사항을 서술하였다. 또한 이러한 방법론을 적용한 예시로서, PISA 2015 자료를 활용하여 학생이 인식한 과학의 도구적 동기 및 즐거움에 대해 국가별 측정동일성 검정 절차를 분석하고 국가별 잠재평균을 추정하였다. 마지막으로 본 연구의 향후 연구 및 의의에 대해 논의하였다.

In multi-group analysis, measurement invariance is a requirement for meaningful comparisons between groups. Multi-group confirmatory factor analysis (MG CFA) has been widely used for group comparisons. However, MG CFA is appropriate for comparisons with a small number of groups and is limited for a large number of groups, in particular, in cross-cultural comparative studies. To overcome the limitation of MG CFA, this study described alternative approaches: multilevel confirmatory factor analysis (ML CFA) and multilevel factor mixture modeling (ML FMM), which are effective for comparing more than 10 groups. In ML CFA, its advantages and disadvantages were described by introducing two models: random intercept models that estimate only intercepts as random effects and random intercept and loading models that estimate intercepts and factor loadings as random effects. Specifically, this study presented theoretical models for the two methods and procedures for testing measurement invariances. In addition, this study discussed advantages of ML CFA, relative to those of MG CFA, and several points that should be considered when applying ML CFA. And, as an example of applying ML CFA, this study conducted latent means analysis across countries for instrumental motivation of science and enjoyment perceived by students using the PISA 2015 data. Finally, implications of this study and future research directions were discussed.

Random Testing(RT) and its derivatives such as Adaptive Random Testing (ART ) are active and important research topics in software testing, which have also a niche in practical settings due to the merits they offers, e.g. fault-detection capacities at low cost, ease of implementation, reliability estimation, facility for automation and so forth. Inspired by these advantages, we elieve the idea behind random testing can be worthwhile and attractive for testing aspectoriented programs since current research on testing of AOP, especially automated has not been adequately performed and is still in infancy. In this paper, we propose a preliminary approach to automated random testing of aspect-oriented programs, which are becoming an important part of software engineering theory and practice. This paper also includes a survey of applicable testing techniques and discussion of established testing methods in both area of Aspect-Oriented Programming (AOP) and Random Testing (RT).

Soft testing is one way of assuring quality of the software. There are many testing techniques to test the software, however random testing is one among the best alternatives, which will generate the test inputs based on some dissemination. It is evaluated theoretically and concluded that the performance of random testing is very impressive when it is compared with partition testing even though it gives good results for testing. Here in this paper we addressed the difference in coverage between random testing and partition testing, here we have taken one example program and written the test cases for that. To test the program we have used automated tools JUnit for partition testing and Randoop for random testing. Once it is over using EclEmma the coverage of test cases is evaluated and proved RT is best.

In real world Testing is the most challenging job in software development. In these days of the information technology epoch, the software has become the lifeline of every human activity. It may not be hyperbole, if someone states that our lives will come to stand-still if every software machine in the universe stops working! When software plays such a crucial role in our lives, it is very important that the software we use should be very high quality and reliability. Even the small gremlin is there in any software that causes huge disaster. These are the examples PATRIOT MISSILE- the patriot missile example shows how a small software bug accumulation of error due to rounding off a number can lead to catastrophe. And ARIANE-5- European Space Agency launched the ariane5 satellite launch vehicle. Exactly 40 seconds after lift-off at an attitude of 3700meters, the launcher devastated and became a ball of fire. This type of mint errors can’t be covered in normal testing’s like Unit Testing, Integration Testing, etc. For that reason RANDOM TESTING is introduced. Random testing is a dominant or a commanding tool in finding low-frequency bugs that are nearly impossible to find using other methods. A Low-frequency bug sometimes leads to system to crash, hence these have to be identified and removed. By using a test, Oracle one may also detect non-crashing errors in the system. Here the objective of work is to study the advantage and various implications of Random Testing, later we will develop one assumed application “Trading system”, to that we will apply random testing to accentuate theoretical and practical implications including variants of Random Testing.

A diagnostic tool for testing homogeneity for random effects is proposed in unbalanced linear mixed model based on score statistic. The finite sample behavior of the test statistic is examined using Monte Carlo experiments examine the chi-square approximation of the test statistic under the null hypothesis.

This paper examines whether stock prices for 16 countries are trend stationary or follow a random walk process using the (Zivot and Andrews, 1992) and (Lumsdaine and Papell, 1997) tests and monthly data (1987:12-2007:04). With one and two structural breaks, the ZA and LP test results provide evidence in favour of random walk hypothesis in 12 and 11 countries, respectively. Thus based on our empirical results the stock market price indices in majority of the countries exhibit a random walk. In addition, the dates of structural break in most cases point to the Asian crisis in the period 1996-1998.

A test based on score statistic is derived for detecting homoscedasticity of errors in unbalanced random effects linear model. A small simulation study is performed to investigate the finite sample behaviour of the test statistic which is known to have an asymptotic chi-square distribution under the null hypothesis.

This paper examines the numerical properties of the nested fixed-point algorithm (NFP) using Monte Carlo experiments in the estimation of Berry, Levinsohn, and Pakes's (1995) random coefficient logit demand model. We find that in speed, convergence and accuracy, nested fixed-point (NFP) approach using Newton's method performs well like a mathematical programming with equilibrium constraints (MPEC) approach adopted by Dube, Fox, and Su (2012).

In this study, we propose two informal hypothesis tests which may be useful in determining the number of trees in a random forest for use in classification. The first test declares that a case is 'easy' if the hypothesis of the equality of probabilities of two most popular classes is rejected. The second test declares that a case is 'hard' if the hypothesis that the relative difference or the margin of victory between the probabilities of two most popular classes is greater than or equal to some small number, say 0.05, is rejected. We propose to continue generating trees until all (or all but a small fraction) of the training cases are declared easy or hard. The advantage of combining the second test along with the first test is that the number of trees required to stop becomes much smaller than the first test only, where all (or all but a small fraction) of the training cases should be declared easy.

랜덤 테스트(RT)는 가능한 입력 도메인에서 임의의 입력 값을 선택하여 테스트 케이스를 생성하고 테스트를 수행하는 기본적인 블랙 박스 테스트 기법이다. 랜덤 테스트의 성능을 향상 시키기 위해서 오류 패턴을 고려한 다양한 Adaptive Random Testing (ART) 알고리즘들이 제안되어 왔다. 그 중 Distance-Based ART (D-ART), Restricted Random Testing (RRT)이 좋은 성능을 보이고 있지만, 수행시간이 너무 느리다는 단점이 있어, 이를 대체할 수 있는 여러 ART 방법들이 제안되고 있다. 그 중, Adaptive Random Testing through Iterative Partitioning (IP-ART)가 가장 좋은 성능과 빠른 수행시간을 보인다. 본 논문에서는 IP-ART 의 성능을 더욱 향상시킬 수 있는 방법을 제안하고, 시뮬레이션을 통하여 향상된 성능을 평가해 보았다.

랜덤 테스트(RT)는 가능한 입력 도메인에서 임의의 입력 값을 선택하여 테스트 케이스를 생성하고 테스트를 수행하는 기본적인 블랙 박스 테스트 기법이다. 랜덤 테스트의 성능을 향상 시키기 위해서 오류 패턴을 고려한 다양한 Adaptive Random Testing (ART) 알고리즘들이 제안되어 왔다. 그 중 Distance-Based ART (D-ART), Restricted Random Testing (RRT)이 좋은 성능을 보이고 있지만, 수행시간이 너무 느리다는 단점이 있어, 이를 대체할 수 있는 여러 ART 방법들이 제안되고 있다. 그 중, Adaptive Random Testing through Iterative Partitioning (IP-ART)가 가장 좋은 성능과 빠른 수행시간을 보인다. 본 논문에서는 IP-ART 의 성능을 더욱 향상시킬 수 있는 방법을 제안하고, 시뮬레이션을 통하여 향상된 성능을 평가해 보았다.

안드로이드 앱의 인텐트 취약점을 테스트하는 인텐트 퍼저에서 에러 확인 방법을 효율적으로 개선한 새로운 아이디어를 제안한다. 인텐트 퍼저는 랜덤 인텐트를 생성하여 앱을 실행한 다음 앱이 비정상 종료되는지 확인하는 테스트 도구이다. 이 논문에서 동일한 에러로 인해 발생한 다수의 비정상 종료 로그들을 하나의 그룹으로 만드는 자동 분류 방법을 제안한다. 테스터는 각 그룹의 대표 로그만 확인하면 된다. 최장 공통 부분 수열을 구하는 알고리즘을 응용하여 이 방법을 설계하였고, 이 방법을 상용 안드로이드 앱 10개에 적용해 실험하였다. 모든 로그를 분석하는 대신 대표 로그를 분석하는 것으로 대체할 수 있음을 확인하였다. 그 결과 분석 대상 로그의 수가 크게 줄었다.