Earticle

본 연구는 "취업진로연구"의 학술 분야 가운데 하나인 ‘취업문화사' 영역이다. 취업문화사 영역은 취 업문화를 사적으로 고찰하고 탐색하는 분야이다. 본 연구는 취업문화의 하위문화인 공채문화 구성요소 들의 공통분모를 형성하고 있는 자기소개서의 시대별 전개 과정과 특징을 비교 분석한 논문이다. 본 연 구의 목적은 1980년대 이래 최근까지 지난 40여 년 동안 변천해온 한국 대기업들의 자기소개서 형태 를 체계적ㆍ종합적으로 고찰한 데 이어 시대별 특징을 비교 분석한 후 이에 대한 시사점을 도출하는 데 그 의의를 두었다. 본 연구의 결론과 시사점을 요약하면 다음과 같다. 첫째, 1980년대 취업시장은 대규모 정기 공채라는 입사 전형의 수월성을 제고 하기 위해 단순한 필 기시험 성적을 중요시하는 가운데 자기소개서의 영향력과 비중은 매우 낮았다. 둘째, 1990년대는 필기 시험 폐지 현상으로 인해 자기소개서가 서류전형에 영향을 미치는 가운데 면접 과정에 비중 있는 참고 자료로 활용되었다. 셋째, IMF 사태 이후부터 2014년 이전까지는 경쟁 지향적인 입사구조로 인해 스펙 중심의 채용제도가 형성된 가운데 지원서의 영향력은 크고 자기소개서의 비중은 낮은 시대였다. 넷째, 2014년 이후의 취업시장은 직무 중심의 채용제도가 정착한 가운데 직무 경험과 직무이해력을 반영하는 자기소개서의 입지가 매우 중요시되고 있다. 한편 공채문화 출현 이후 전개되어온 자기소개서의 역사가 주는 시사점은 그동안의 경쟁 지향적인 인재 확보를 위한 수단보다는 스펙이 낮거나 혹은 잠재능력을 가진 구직자들을 위한 한층 진화된 자기 소개서의 개발이 필요한 시점이라는 것이다. 직무제안서, 사업기획서, 사업 아이템 공모전, 선진 직무개 발서가 대안이 되고 있다. 이는 스펙은 입사하는 순간까지만 적용되고 이후는 그것이 성과와 생산성을 보장할 수 없기 때문이다. 따라서 이런 대안들은 직무역량과 사업이해력이 높을수록 조직 및 기업경쟁 력을 배가시킬 수 있다는 판단에 기인한 평가방식이다.

This study is a thesis in the field of the history of employment culture. The purpose of this study is to derive implications after comprehensively comparing and analyzing the transition process and characteristics of Korean large companies'cover letter forms from the 1980s to 2022. The conclusions and implications of this study are as follows. First, in the 1980s, the employment market valued simple written test scores in order to increase the effectiveness of the employment test of large-scale regular recruitment, and the influence and proportion of cover letters were low. Second, in the 1990s, cover letters influenced document screening due to the abolition of written tests, and were used as reference materials that weighed on the interview process. Third, from the IMF crisis to 2014, a spec-oriented recruitment system was formed due to the competitive employment structure, and the proportion of applications was high and the proportion of cover letters was low. Fourth, in the employment market after 2014, while the job-oriented recruitment system has been established, cover letters that reflect job experience are becoming important. On the other hand, the implication of the history of cover letters that have been developed since the emergence of the public recruitment culture is that it is necessary to develop more advanced cover letters for applicants with low specifications or potential rather than means to secure competitive talented people. Job proposals, business plans, business item contests, and advanced job development documents are alternatives. This is because the specification is applied only until the moment of joining the company, and after that, it cannot guarantee performance and productivity. Therefore, these alternatives are evaluation methods based on the judgment that the higher the understanding of job competency and business, the stronger the organizational and corporate competitiveness can be.

Face-to-face interview format has been widely used to elicit English language samples in oral proficiency testing. It is desirable to replicate target language use situations in oral testing, and the direct testing format is believed to assess more authentic and interactive language abilities. However, it has been argued that speech samples from unstructured face-to-face interviews are quite different from those in natural communication settings. The purpose of this study was to understand the construct to be assessed through an unstructured face-to-face interview, which was adopted to an English speaking contest. Data from the English speaking contest were analyzed in terms of how they were different from the characteristics of naturally occurring conversations. It was found that the test construct in the English speaking contest did not reflect the features of natural conversation specifically in the areas of turn-taking, adjacency pairs, and topic nomination due to asymmetrical power relations between interviewer and interviewees. Therefore, it was suggested that we need to incorporate diverse discourse-based approaches into current speaking skill assessments, which can interpret spoken language data in many ways.

This study was conducted to examine students' and teachers' attitudes to a Computerized Oral Test (COT) and a Face-To-Face Interview (FTFI). The results of this study showed that the students had significantly more positive attitudes to the COT in terms of preparation time, nervousness, tiredness, and fairness, but more students tended to prefer the FTFI primarily because it better approximated real-time interaction and conversation. On the other hand, the teachers clearly showed more favorable attitudes to the COT, especially with regard to various question types, fairness, and practicality despite some perceived weaknesses of the COT; it poorly approximated real-life interaction and therefore felt less valid. On the whole, the teachers' attitudes and reactions to the COT were more positive than those of the students.

본 연구는 화병환자의 진단을 위해 선행연구들과 임상사례보고서를 토대로 화병에 대한 진단준거를 만들고, 이를 기초로 하여 화병면담검사(Hwa-Byung Diagnostic Interview Schedule, HBDIS)를 제작하였으며, 이에 대하여 신뢰도 및 타당도를 검증하였다. 이를 위하여 K대학 신경정신과 우울증 센터를 방문한 환자 55명과 K대학 한방병원 화병클리닉을 방문한 환자 26명을 대상으로 화병면담검사를 실시하였다. 화병면담검사의 검사-재검사 신뢰도를 산출하기 위해 1개월 후 재검이 가능한 28명에 대해 두 명의 임상가가 화병면담검사를 재실시하고, 검사-재검사 신뢰도를 살펴본 결과 .82로 나타났다. 다음 화병면담검사의 타당도를 검증하기 위해 환자의 차트를 보고 두 명의 임상가가 내린 임상진단과 화병면담검사를 통해 내린 진단의 일치도를 살펴보았다. 그 결과 전반적 일치도(Overall agreement)는 .85로 나왔고, 진단의 민감도(Sensitivity)는 .83, 특이도(Specificity)는 .88로 나타났다. 이와 같은 결과는 화병면담검사가 화병을 진단하는데 신뢰롭고 타당한 도구로 사용될 수 있음을 시사하였다.

Hwa-Byung has long been known as a form of psychogenic illness among lay people in Korea and recently listed as a culture-bound syndrome of Korea in DSM-IV. However, there has been neither diagnostic criteria nor diagnostic instrument, which rendered empirical studies liable to subjective biases. The purpose of this study was to establish diagnostic criteria of Hwa-Byung, develop a structured diagnostic interview schedule for Hwa-Byung, and examine its reliability and validity. Participants were 55 patients visiting the depression clinic and 26 patients visiting the Hwa-Byung clinic. The diagnostic criteria was first determined on the basis of previous research findings and clinical case studies. Then the initial items of Hwa-Byung Diagnostic Interview Schedule (HBDIS) were selected to obtain information necessary to diagnose Hwa-Byung. This initial interview schedule was administered to 31 patients, which results were discussed and reviewed by all the authors. After some additional modifications based upon interviewers' comments, the present form of the instrument was completed and conducted to 81 patients. Independent clinical diagnoses for each patient were made by psychiatrists, psychiatrists of oriental medicine and clinical psychologists, which rate of agreement was demonstrated to be .84. The test-retest reliability of the HBDIS was examined for 28 patients and calculated to be .82. The validity of the HBDIS was evaluated on the extent to which HBDIS diagnosis assigned to 81 patients agreed with diagnosis determined by clinicians. As a result, Overall agreement was .85, Sensitivity was .83, and Specificity was .88. The kappa coefficient and other validity indicators were in the acceptable range (.70 - .88), indicating that the HBDIS was valid. Limitations and applications of the HBDIS were discussed.

본 연구의 목적은 대학생이 지각하는 토익스피킹 시험의 교육적 효용성에 대한 인식이 취업 영어면접 준비태도에 미치는 영향과 이 과정에서 학습전략이 수행하는 매개효과를 검증하는 데 있다. 이를 위해 서울·경기지역 소재 4년제 대학 4곳의 재학생을 대상으로 설문조사를 실시하였으며, 총 250부의 설문지 중 249부를 최종분석에 활용하였다. 매개효과 검증은 Baron과 Kenny(1986)가 제시한 3단계 절차를 적용하였다. 분석 결과, 토익스피킹 시험의 교육적 효용성에 대한 인식은 학습전략에 유의한 정(+)의 영향을 미쳤으며, 학습전략은 취업영어면접 준비 태도에 긍정적인 영향을 미치는 것으로 나타났다. 또한, 토익스피킹 시험의 효용성 인식은 취업영어면접 준비 태도에 직접적인 영향을 미칠 뿐만 아니라, 학습전략을 매개로 간접적인 영향도 미치는 부분 매개효과가 확인되었다. 이러한 결과는 토익스피킹 시험이 단순한 평가 도구를 넘어, 대학생의 전략적 영어 학습을 촉진하고 취업 영어면접 준비 태도 형성을 강화하는 교육적 자원으로 활용될 수 있음을 시사한다.

The purpose of this study was to examine the effect of university students’ perceived educational usefulness of the TOEIC Speaking Test on their English Job interview preparation attitude and to verify the mediating role of learning strategies. A survey was conducted with students enrolled in four four-year universities located in Seoul and Gyeonggi Province, and 249 valid responses out of 250 questionnaires were used for the final analysis. The mediating effect was tested using the three-step procedure proposed by Baron and Kenny (1986). The results indicated that perceived educational usefulness of the TOEIC Speaking Test had a significant positive effect on learning strategies, and learning strategies also positively influenced employment English Job interview preparation attitude. In addition, perceived educational usefulness directly affected English Job interview preparation attitude and also indirectly influenced it through learning strategies, confirming a partial mediating effect. This study provides empirical evidence that the TOEIC Speaking Test can be utilized not merely as an assessment tool but as an educational resource that promotes strategic learning and strengthens English Job interview preparation attitudes in university English education.

Despite positive influence of feedback on students’ learning in language assessment (Saito & Lyster, 2012), the nature and impact of raters’ feedback in language assessment remains underexplored, especially its influence on English speaking class students and instructors. Situated in an institutional oral proficiency interview (OPI) test in a university in the U.S., this study aimed to identify the varying needs of different stakeholders involved in the speaking assessment, including raters, students and instructors. The OPI is a high-stakes test in this context, serving as a decisive factor for allocating teaching responsibilities and determining financial assistance for prospective international graduate students within the department. Utilizing a mixed-methods approach, the research involved focus group interviews with five OPI raters and two instructors of English speaking classes, respectively. A questionnaire was completed by twenty-two international graduate students preparing for the test. The findings revealed that all stakeholders recognized significance of raters’ feedback whereas their perceptions regarding the quality and diagnostic utility of raters’ feedback varied. This study concluded by offering suggestions to improve the quality of raters’ feedback for English speaking tests, both in the studied institution and in broader contexts. The suggestions provided a practical foundation for developing an online rating system for the speaking test in future studies.

심리생리검사는 사람이 거짓말을 하게 되면 생기는 죄책감이나 탄로의 우려 등과 같은 심리상태가 형성될 때 동반되는 생리적인 변화를 측정, 분석함으로써 진술의 진위 여부를 판단하는 심리분석 기법을 말한다. 국내에서는 주로 범죄 피의자에 대하여 심리생리검사를 하고 있으며 이를 통해 중요사건이 해결된 사례가 많이 있다. 현재까지 심리생리검사의 정확도, 증거능력 등에 관하여 많은 연구논문이 발표되면서 심리생리검사는 수사과정에서 중요한 한 축을 이루는 과학적 검사기법으로 자리를 잡게 되었다. 기본적으로 심리생리검사는 3단계의 절차로 구성되어 있다. 즉 검사 전 면담, 검사, 검사 후 면담이 그것이다. 그러나 그 동안 국내 대부분의 심리생리검사는 검사 후 면담을 실행하지 않았다. 검사 후 면담을 지양했던 주요 이유는 심리생리검사관에게 중립적인 감정관으로서의 판단을 보장하기 위한 방편이었다. 심리생리검사관은 피검사자에게 명확한 거짓이 나타난 경우에도 검사 후 면담의 절차를 진행하지 않았다. 그런 이유로 심리생리검사관은 피검사자로부터 자백이나 추가 진술을 통해 검사의 정확도를 검증할 수 없었다. 기존의 연구는 피검사자에 대한 검찰의 기소 또는 법원의 유죄 판결로 심리생리검사의 정확도를 추정하였다. 그런데 2016년 한국의 대검찰청 심리분석실이 미국 폴리그래프협회로부터 정식 교육기관으로 인증을 받게 되면서 검사 후 면담제도를 시범 실시하게 되었다. 그 결과 다수의 피검사자로부터 심리생리검사 결과에 부합하는 진술을 받음으로써 검사의 정확도까지 검증해 볼 수 있었다. 이에 본 연구는 심리생리검사의 중요 과정인 검사 후 면담의 실제 사례 및 효용성을 검증한 것에 그 의의가 있다.

A Polygraph test refers to a psychological analysis technique that determines the validity of statements by analyzing and measuring the physiological changes caused by guilty feeling or fear of disclosure when a examinee is lying. In Korea, polygraph tests are used mainly to measure the credibility of criminal suspect’ statement and have been considered useful in solving criminal cases. Many research papers have been reported about the accuracy of polygraph tests and admissibility of evidence. As a result, polygraph test have been acknowledged as one of the important scientific inspection techniques. Basically, polygraph test consist of three steps: pre-test interview, actual test, post-test interview. Most of polygraph tests in Korea, however, do not include the post-test interview. The main reason for the skipping of the post-test interview was to guarantee the examiner can decide as a neutral analyst. Therefore examiners in Korea skip the stage of post-test interview even if the examinee’s physiological responses were apparently observed. For that reason, the examiner could not calculated and verified the accuracy of polygraph tests by means of getting a confession or additional statements on a issue of concern from the examinee after the tests. Previous studies have estimated the accuracy of polygraph tests by prosecutors’ indictment of the examinees or the conviction of the court on them. In 2016, Psychological Forensic Center of Supreme Prosecutors' Office in Korea was confirmed as a formal educational institution by the American Polygraph Association, and we conducted a demonstration of the post-test interview. As a result, we found a number of significant cases in which additional meaningful statements were acquired during post-test interview. And we verified the accuracy of the polygraph test by comparing the contents of additional statements from the examinees and the results of polygraph tests. This study is meaningful in that a post-test interview which plays an important role in a polygraph test was applied in real-life cases and verified the efficiency of that stage.

The purpose of this study is to investigate whether classroom-based interview test of Korean proficiency can reflect authentic ways people speak in non-test situations. This study analyzed interview data collected from 11 KFL (Korean as a Foreign Language) students in Korea. It was found that the interlocutor-interviewee discourse is highly asymmetrical in terms of features of turn distribution, dominance, mutual contingency and goal-orientation. The fixed role relationship (a native speaker of Korean and nonnative speakers, or teacher and student) in the one-to-one test format makes it difficult for the interviewees to escape this asymmetry. The finding from this study implicates that various interaction pattern should be used, and that interactional competence need to be considered in a speaking test of Korean.

The purposes of this study is to identify what variables are important in increasing the validity and reliability of oral interview test from the learners’ perspectives. Factor analysis was employed to identify what variables were involved, and analysis of variance was used to compare the degree of importance that students and instructors rated on these variables. According to the results of this study, variables related to testing methods, testing settings, standardization procedures, characteristics of students, students’ needs and goal, and students’ linguistic and social backgrounds were found to be influencing the validity and reliability of oral interview test. Especially, it was found that the variables related to students and testing environment strongly influenced the satisfaction that students felt about the test itself. And distinctive differences on the degrees of importance students and instructors rated on these variables were found. From students’ perspectives, these variables are likely to limit the full expression of their language abilities. Thus, instructors need to consider these variables carefully in the process of oral interview test to increase the validity and reliability of test.

본고는 한국어 고급 학습자의 인터뷰 평가 담화를 분석하여 응결 장치 사용을 통한 담화 응집성을 고찰하였다. 양적인 분석과 질적인 분석을 통해 응결 장치인 연결어미와 접속부사, 지시․대용 표현의 사용 양상과 사용 적절성을 분석하였고, 학습자 L1 변인에 따른 사용 양상의 특징을 분석하였다. 담화의 응결 장치 중에서 연결어미, 접속부사, 그리고 지시․대용 표현에 한정하여 분석한 결과, 응결 장치의 사용 양상이 상이하게 나타났다. 일본인 학습자는 구어 담화에서 접속과 지시․대용의 응집성을 실현하고 있었다. 반면에 몽골인 학습자는 지시․대용 표현의 오류 사용과 간투사의 잦은 사용으로 인해 구어 담화에서 지시․대용의 응집성을 실현하지 못하였고, 중국인 학습자는 응결 장치의 저빈도 사용과 높은 오류율, 간투사의 잦은 사용 때문에 응집성을 실현하지 못하였다. 구어 담화는 화자와 청자의 상호작용이 있으며, 이해와 발화가 자동적으로 이루어져야 원활한 의사소통이 이루어지므로 청자의 이해를 돕고 발화 맥락을 알 수 있게 해 주는 응결 장치가 매우 중요한 역할을 하게 된다. 화자는 응결 장치를 사용함으로써 담화 내용을 논리적으로 연결시켜 응집성을 강화할 수 있고, 청자는 담화의 맥락을 논리적으로 유추할 수 있게 된다. 응결 장치의 사용 양상과 L1 변인에 따른 학습자 특징을 다룬 본고는 담화 응집성을 보완해 줄 수 있는 교수․학습의 기초 연구가 될 수 있을 것이다.

This study analyzes interview test given by advanced Korean language learners to examine discourse cohesion shown upon using cohesive device. It focuses on the use of connective ending, conjunctive adverb, and reference?substitution, these as a cohesive device, in terms of their application, appropriacy and use patterns, in particular, based on the variable L1 of learners. The study is measured with quantitative analysis and with qualitative analysis. The analysis is confined to the use of connective ending, conjunctive adverb and reference?substitution among cohesive devices of discourse. It is noteworthy that high error rate and low using rate of the cohesive devices, and using the high-frequency of filler in Chinese students interfere cohesion and high error rate of reference?substitution and using the high-frequency of filler in Mongolian students interfere cohesion. Spoken Discourse is based on the interaction between speaker and hearer, and automatical input and output. In the communication cohesive devices help the understanding of the hearer and become a very important role in spoken discourse. The use of cohesive devices in spoken discourse helps logical connection of discourse, thus reinforcing discourse cohesion and enabling a communication partner to make logical inference about context of the discourse. In this regard, the study is a basic research to improve teaching and learning methods in accordance with the variable L1-based, different type of learners.

본 연구는 의사소통 전략에 관한 논의에서 ‘어휘’가 가지는 위상을 재고찰하고 한국어 학습자들의 어휘 사용 전략을 면밀하게 관찰하여 제시하였으며 이를 통해 어휘 사용의 전략이 어휘 유창성의 한 하위 범주로 설정될 수 있는지에 대한 가능성도 타진해 보고자 하였다. 본 연구에서는 학습자가 산출한 문어, 구어 자료를 중심으로 어휘 사용 전략의 예를 살피고 이를 범주화하였다. 그 결과 학습자들은 ‘문제 해결 전략’과 ‘어휘 운용 전략’을 활용하여 의사소통 수행 중에 나타나는 문제 상황을 해결해 나가고 있었으며 한편으로는 전달하고자 하는 의미와 형식에 적절한 어휘를 선택하고 사용함으로써 의사소통의 효율성을 증대시키고자 노력하고 있었다. 이러한 학습자들의 노력은 궁극적으로 의사소통 수행의 질(質)을 향상시키고 있었으며 이는 ‘수행’을 중심으로 하는 어휘 유창성 논의와도 연관지어 살펴볼 수 있었다. 따라서 본고는 어휘 사용과 관련한 전략적 능력이 어휘 유창성 정도를 판단하기 위한 기준으로 활용될 수 있다는 입장을 취하고자 한다.

The purpose of this paper is to perform discourse analysis of intermediate and advanced students’ writing and interview test, in order to rethink strategies competence in the vocabulary domain. In addition, this study aims to explore the vocabulary use strategies that could be established as a subcategory of lexical fluency. For this study, written and spoken discourse analysis followed finding examples and cataloging of vocabulary use strategies. The results show that students use ‘problem-solving strategies’ and ‘vocabulary operation strategies’ to solve the many problems that arise during communication and to promote communicative efficiency by selecting and using appropriate vocabulary for communicative situational contexts. The students’ efforts ultimately improve the quality of communication performance and could be attached to lexical fluency. Therefore, this study argues that vocabulary use strategies could be used as a basis of evaluation for estimating a degree of a lexical fluency.

국내에서 2000년 이후 인터넷 중독에 대한 상담이 광범위하게 시행되고 있으나, 인터넷 중독 상담의 기록이 표준화된 방식으로 운용되지 못하고 있어 상담서비스의 질적 개선이나 정책 수립을 위해 활용되지 못하고 있는 실정이다. 따라서 본 연구는 체계화된 절차에 따라 인터넷 중독 상담 기록을 위한 면접 및 심리평가 배터리를 제안하고자 하였다. 먼저, 우울, 주의력-충동성, 상담만족도 변인은 임상 및 상담 장면에서 널리 쓰이고 광범위한 타당화가 이루어진 BDI, CDI, Corner’s ADHD scale, 작업동맹척도, CSQ-8의 사용을 제안하였다. 둘째, 타당도가 충분히 검증되지 못한 것으로 조사된 부모자녀-의사소통, 대인관계 유능성, 학교생활 적응, 또래관계 만족도 척도는 문헌조사와 기존의 척도에 기초하여 수정 보완하였다. 수정 보완된 척도들에 대한 신뢰도와 타당도 분석을 확인한 결과 양호한 것으로 나타났다. 또한, 한국정보화진흥원의 인터넷 중독 상담센터의 자료를 분석한 결과, 게임 중독 비율이 높은 것으로 확인되어, DSM-5의 게임 중독 진단에 기반한 면접 질문지를 추가하였다. 본 연구에서 제안한 면접 및 심리평가 배터리가 인터넷 중독의 상담 기록에 활용된다면, 인터넷 중독자에 대한 이해를 높이고 효율적으로 치료하고 관련된 연구를 보다 체계적으로 수행하는데 도움을 줄 것이라 기대한다.

Although the internet addiction counseling is broadly practiced and related data are greatly cumulated, such efforts are not utilized for qualitative improvement or policy making of the counseling service. The reason is that the records of internet addiction counseling has not been standardized yet. Therefore, this study is done to propose a interview and psychometric battery assessing internet addiction. The counseling record form is composed of items measuring demographic information, tendency for internet addiction, features associated with internet game addiction, depression, attention-impulsivity, working alliance, client satisfaction, interpersonal competency, parent-adolescent communication, and school life adjustment. In our study, Questionnaires which validation and reliability were not repeatedly confirmed in previous research, were corrected and complemented: those are about extra-personal qualities including interpersonal competency, peer relationship satisfaction, parent-adolescent communication, and school life adjustment. The current test battery for internet addiction was composed by integrating theoretical background and clinical practice results in this research; accordingly it could be used validly in other research or real counseling settings.

본 연구는 2020년 공군사관학교 선발면접을 실시한 227명의 지원자를 대상으로 AI 면접의 타당도를 검증하는데 목적이 있다. 이를 위해 지원자의 AI 면접과 대면면접 결과의 연관성과 이에 따른 생도 성취도를 각각 분석하였다. 연구결과에 의하면, 첫째, 면접 최종판정인 AI 면접의 추천/비추천 결과에 따라 대면면접의 합격/불합격 결과가 일치하지 않아 두 면접 간 차이가 나타났다. 둘째, AI 면접과 대면면접 모두 생도의 학업성적 및 종합성적과 유의한 상관관계가 나타났으나, 대면면접이 AI면접보다 더 큰 관련성이 있었다. 셋째, AI 면접 판정인 추천/비추천 결과에 따라 향후 생도로서 성취도 차이는 나타나지 않았다. 따라서 사관학교 AI 면접은 생도의 특수성을 반영한 커스터마이징 방식의 도입이 요구된다.

This study verified the validity of AI interview conducted on 227 people who wereinterviewed for the selection of Air Force Academy in 2020. The results of the study are as follows: First, the results of the interview were not consistent with the results of the interview,which is the final decision of the interview, and the results of the interview were not consistent. Second, both AI interviews and face-to-face interviews showed a significant correlation withcadet's academic and comprehensive grades, but face-to-face interviews were more related thanAI interviews. Third, there was no difference in the achievement of cadets in the futureaccording to the results of recommendation/ non-recommendation, which is the final judgmentof AI interviews.

The purpose of this study was to investigate inter- and intra- rater reliability in aninterview and a computerized oral test. It was also examined whether ratercharacteristics influenced on their reliability and biases, and finally the scores of bothtests were compared with those of the Versant test using an automated computer ratingsystem. For the study, the data from 21 Korean university students and 18 Korean ornative speakers of English raters with various characteristics were collected. Some ofthe main findings from the study were as follows. First, rater severity was significantlydifferent in each test, but each rater consistently graded on both tests suggesting lowerinter-rater reliability and higher intra-rater reliability. Secondly, rater severity wasimpacted by the rater characteristics such as mother tongue, gender, age, and major. Lastly, there existed a positive correlation among the scores of the three tests,indicating that the scores of human beings and computers are strongly related.