Ehime University Institutional Repository

Ehime University (愛媛大学)博士(農学)thesi

Mulyana, Hadipernata

 
 
（第５号様式）  
                        学位論文審査の結果の要旨 
 
    氏    名       Mulyana Hadipernata 
審 査 委 員 
       
 
         論  文  名 
  Effect of ᴅ-Ketohexoses on Rheological Properties of Chicken Gel Products  
  （鶏肉および鶏卵のゲル状食品の物性に及ぼすＤ-ケトヘキソースの添加効果と 
   その機構に関する研究） 
 
審査結果の要旨  
 
 糖類の過剰摂取は生活習慣病の発症リスクが高まることから、糖類を低減した加工食品が増えてきて
いる。しかし、糖類には澱粉の老化抑制や細菌の増殖抑制など甘さ以外にも様々な働きがあることから
加工食品には欠かせない食品素材である。近年、糖類の一つである D-プシコース（Psi）に抗糖尿病作
用、脂肪蓄積抑制作用などが見出され注目を集めている。Psi は D-フルクトース（Fru）の異性体である
が Fru と違いカロリーがほぼゼロであることに加え、カスタードプリン、メレンゲ、求肥などの加工食
品に添加すると優れた物性及び品質劣化抑制効果を示すことから、食品素材としても注目されている。
しかし、Psi に冷凍食品の品質劣化抑制効果があるかどうかは明らかとなっていない。本研究では、Psi
をはじめとする D-ケトヘキソース（DKH）を鶏の胸肉や卵を主原料とするタンパク質加熱ゲル食品に
使用し、作製したゲル状食品の物性及び冷凍保存したときのゲル物性の変化を調べることで、DKH に冷
凍保存によるゲル物性の劣化を抑制する働きがあることを明らかにした。 
 
 まず、Psi を鶏胸肉ソーセージ（CBS）に使用した。胸肉原料に対して添加する 2.5%Suc を Psi で 30％、
70％、100％置換して CBS を作製した。各 CBS の破断強度、クリープ測定、保水性測定を行ったところ、
Psi100％使用 CBS の物性は Suc100％使用のものと大差なかったが、保水性は約 30％高かった。CBS を
はじめ多くのタンパク質ゲル状食品は冷凍貯蔵すると粘弾性の低下や離水を引き起こす。そこで、各
CBS を-20℃で 90 日間凍結した後、解凍し物性を調べた。100％Psi 使用 CBS は 100％Suc 使用 CBS と比
べ弾性回復率が高かったことから、Psi にはソーセージの弾性回復率の低下抑制効果があることが分か
った。この効果は Psi がもつ高い保水力によるものと推定された。以上のことより Psi は Suc よりも冷
凍保存によるゲル物性の劣化を抑制する効果が高いと結論づけた。 
 
 
 
 次に、全種類の DKH、すなわち Psi、Fru、D-ソルボース（Sor）、D-タガトース（Tag）を鶏卵白の加
熱ゲルに使用した。卵白に 10％DKH を加えて加熱した DKH 添加ゲルとスクロース（Suc）、D-グルコ
ース（Glc）、トレハロース（Tre）を添加した加熱ゲルの物性を比較した。DKH 添加卵白ゲルは、Suc、
Glc、Tre 添加のものよりも、ゲルの破断強度と破断歪み率が有意に高かったことから、壊れにくいゲル
ネットワークを形成していることがわかった。この壊れにくいゲルネットワークには、DKH が加熱工程
で卵白タンパク質とメイラード反応を引き起こし、その結果生じるタンパク質間の架橋構造が関与して
いることが推察された。また、DHK 添加ゲルは、Suc、Glc、Tre 添加のものの約 2 倍の粘性率値を示し
たことから、DHK は卵白タンパク質ゲルの粘性を著しく高めることが分かった。次に、凍結融解を１～
３回繰り返すことにより、卵白ゲルの劣化を強制的に引き起こさせて、DKH の冷凍貯蔵による品質劣化
抑制効果を評価した。その結果、凍結融解後のゲルの弾性係数、粘性率、破断強度、破断歪み率は、い
ずれも DKH 添加の卵白ゲルの方が、Suc、Glc、Tre 添加のものよりも高かった。これらの結果は、DHK
添加卵白ゲルが Suc や Tre を添加したものよりも冷凍保存による物性の劣化を起こしにくいことを示し
ている。このことより４種類の DKH はいずれも冷凍保存した卵白ゲルの品質保持において、Suc や Tre
よりも優れた糖であることを明らかにした。 
 
 4 種類の DKH を卵白ゲルと同様に全卵の加熱ゲルに使用した。DKH 添加全卵ゲルの弾性率、粘性率
は Suc、Glc、Tre 添加のものとほとんど同じ値であったが、ゲルの破断強度、破断歪み率は有意に高か
った。このことから、DKH は全卵ゲルに使用してもゲルを壊れにくくするといえるが、その効果は卵白
ゲルのときほど大きくはなかった。卵白ゲルと全卵ゲルでの DKH の効果の違いは全卵に含まれる脂質
成分が DKH の働きを弱めたためと考えられた。凍結融解による全卵ゲルの物性変化については卵白ゲ
ルのときほど DKH に著しい優位性は示されなかったものの、弾性率、破断強度については Suc、Glc、
Tre を添加したゲルよりも DKH 添加ゲルの方が高かった。このことより全卵ゲルにおいても DHK は冷
凍耐性を示すことが示された。 
 
 本研究では、DKH の効果としてタンパク質ゲルの物性に着目して調べたが、もう一つ Psi の食品組織
への浸透性についてモデル食品としてジャガイモを使って調べた。その結果、Psi は Suc よりも浸透速
度が高いことがわかった。このことは Psi が Suc よりも浸漬しやすいということを示しており、浸漬液
の食品組織への浸漬時間を短縮できることを示唆している。 
 
 本論文は、Psi をはじめとした DKH が Tre や Suc よりもタンパク質加熱ゲル食品の物性改善に優れて
いることを明らかにした。また、DKH は冷凍保存による品質劣化の抑制効果においても、Tre や Suc よ
りも優れていることを示した。このように申請者が見出した研究結果は、食品素材としての Psi や Sor
の新たな用途開発につながるものであり、博士論文として高く評価できる。 
 
 本論文に関する公開審査会は、平成 28 年８月６日に香川大学農学部で開催され、論文発表と質疑応
答が行われた。続いて開催された学位論文審査委員会において本論文の内容について慎重に審査を行っ
た結果、審査委員全員一致して博士（農学）の学位を授与するに値するものと判定した。 
 


English

Effect of ?-Ketohexoses on Rheological Properties of Chicken Gel Products

Ehime University (愛媛大学)博士(学術)thesi

SHINDE, DHANASHRI BALASO

Development of a Multi-operation System for an Intelligent Greenhouse

上野, 進

Japanese

近現代の四国遍路研究によせて

中川, 未来

植民地台湾の四国八十八ヶ所写し霊場

Ehime University (愛媛大学)博士(工学)doctoral thesi

志田, 洋

SHIDA, Hiroshi

安全性と信頼性を考慮した鉄道信号設備のライフサイクルコスト低減に関する研究

REINELT, Rudolf

German

(Deutsch)fremdsprachenunterricht in Strangen I

寺谷, 亮司

愛媛大学地域創成研究センター活動報告 ―平成 27 年度（2015 年 4 月~2016 年 3 月）― 
 
 2015 年における地域創成研究センター教員は，センター長・寺谷亮司教授，和田寿博教授，佐藤亮
子准教授，川岡勉教授（教育学部），松野尾裕教授（教育学部），野崎賢也准教授（法文学部）の 6
名であり，前 3 者が当センター専任教員である。2015 年度は，下記 2 部門において，多くの活動を実
施した。 
 
１．学術文化創造部門 
 各センター教員は，地域に根ざした研究や教育効果を意識した学生との協同調査を実施している。
前者には，「松山空襲など戦争体験記録プロジェクト」などがあり，後者の成果の一部は，高知市実
習調査報告など，本年度に刊行した地域創成研究年報第 11 号にも掲載されている。 
 共同研究の推進・募集と支援に関しては，「えひめ 311」，「愛媛グローバルネットワーク」など
の NPO との交流活動を実施し，また「四国遍路と世界の巡礼研究会」などの講演会を支援・後援した。 
 地域の文化・芸術事業の推進としては，3 月 8 日に，昨年度の第 1 弾に続くシンポジウム第 2 弾「注
目の発酵茶：どう活かしどう伝える？―産地の連携した取り組みを考える―」を開催し，全国から 100
名を超える参加者を得て，NHK などからの取材を受け，同日のニュース番組で放映された。 
 
２．地域政策部門 
 同部門におけるシンクタンク機能の推進に関しては，松山市との「道後温泉活性化共同事業」，「道
後温泉活性化懇談会」，「松山市清水地区まちづくり協議会」，「松山市中心市街地活性化事業」，
「松山市中小企業振興基本条例検討委員会」，さらに「宇和島市津島町御槇地区集落コミュニティ支
援事業」，「上島町魚島地域元気アッププロジェクト」，「今治市中心市街地再生基本計画策定審議
会」を継続・実施した。 
 また，和田寿博教員を中心に，経済産業省「平成 27 年度産学連携サービス経営人材育成事業」に採
択され，下記の「産学官連携による観光サービス産業の経営管理を担う人材育成事業」を実施し，12
名の修了者を得た。 
 ・試行講座・観光イブニングセミナー計 6 回（①10 月 14 日：講師・伊東豊氏（中小企業基盤整備
機構四国本部松山オフィス）「地域観光産業の活性化と人材育成」，②11 月 25 日：講師・一色昭造
氏（石崎汽船株式会社社長）「地域観光と瀬戸内の交通」，③12 月 2 日：講師・宮内雅史氏（いよぎ
ん地域経済研究センター主任研究員）「道後温泉観光の歴史と現状」，④1 月 13 日：講師・宮崎光彦
氏（道後温泉宝荘ホテル株式会社社長）「道後旅館業とサービス経営」，⑤1 月 20 日：講師・佐藤誠
氏（熊本大学名誉教授／元北海道大学観光高等研究センター特任教授）「持続的なツーリズムと観光
サービス」，⑥1 月 27 日：講師・和田寿博氏（愛媛大学地域創成研究センター兼法文学部教授）「ま
とめ」） 
 ・観光産業振興研究会計 6 回（①10 月 28 日：報告者・河内広志氏（道後プリンスホテル社長）「『朧
月夜』の経営戦略」，②11 月 11 日：報告者・森健次氏「くし秀の地産地消とオーナー型経営」，③
11 月 18 日：報告者・小田美恵氏（松山市中島 民宿宝山亭）「離島民宿と地域活性化」，④11 月 30
日：報告者・大倉忠司氏（鳥貴族社長）「鳥貴族の全国展開と経営課題」，⑤12 月 9 日：報告者・奥
村武久氏（大和屋会長）「観光立国と地域観光政策」，⑥12 月 16 日：報告者・藤渕利通氏（有限会
社内子ワイナリー社長）「内子物産品の経営」） 
 ・産学官連携コンソーシアムの設置，同懇談会の実施 
 地域貢献事業では，「mit まちなか大学」を開催し，市民に向けた「知」の発信に努めた。今年度は，
第 1 回「商店街ポスター展を通して見えてきた三津の魅力」（4 月 18 日，n’s kitchen**&labo 裏倉庫），
第 2 回「お酒を造ってくれる日本酒酵母 －EK1（愛媛酵母）を中心に－」（12 月 2 日，蔵元屋），
第 3 回「ワイン文化を考える－内子ワインと島根わいん・奥出雲ワインを中心に－」（1 月 27 日，清
水ふれあいセンター）を内容とする講演会を実施した。 
 松山市との共同事業「地域づくり支援セミナー」では，防災をテーマとして，第 1 回（6 月 3 日～
20 日，土砂災害・津波のリスクがある地域の減災－粟井地区－）および第 2 回（10 月 29 日～11 月
14 日，密集市街地の減災－味酒地区－）を実施した。同事業の受講生は 62 名，うち修了書授与者は
55 名であり，地域人材の養成に貢献した。                   （寺谷 亮司） 
地域創成研究年報第１１号（2016）
79


愛媛大学地域創成研究センター活動報告 : 平成27年度(2015年4月～2016年3月)

向, 平和

隅田, 学

中本, 剛

大橋, 淳史

熊谷, 隆至

日詰, 雅博

中村, 依子

佐野, 栄

理科観察・実験の指導力育成に向けた取り組み

Mark Duane STAFFORD,

Introduction
The English Education Center（EEC）at Ehime
University, founded in 2001, began developing a com-
mon test program for the university’s required general
education English courses during the 2009－2010
academic year. The goals of the common test program
were to establish minimum English proficiency levels for
students to attain during their freshmen-year, to assess
whether the goals and objectives of the program were
being met, and whether students demonstrated improve-
ment in their abilities over four semester-long skill-based
English courses.
EEC courses were split into separate listening,
speaking, reading, and writing courses in2007. Listen-
ing and speaking are offered in the first semester and
reading and writing in the second. Common tests for
the speaking and writing courses are based on rubrics
developed by members of the EEC, whose development
underwent a number of iterations before being finalized
into the current form which has remained in use for the
past seven years.
The development of the listening and reading
common tests, on the other hand, has been decidedly
less expedient for a variety of reasons. First of all,
substantial time and effort has been required in
developing the current200 listening test questions1）and
120 reading test questions compared to the rubric-based
tests. While scoring descriptors within the rubrics can
be quickly and easily revised on a single form, each
multiple-choice-based question must be piloted, examined
for effectiveness（by employing classical test theory
［CTT］item analysis statistics）（Brown,1996）, revised
based on such data, and re-piloted, re-examined, and
revised again in an ongoing process until desired results
are obtained. In addition, each individual test question
must be evaluated with regard to its contribution to the
overall difficulty of the test ― i. e. the mean score. Of
the EEC’s listening and reading A, B, C, and D sets, the
A forms have gone through the developmental process
at least six times and are therefore the most
sophisticated in terms of statistical quality. The A sets
still need to undergo a few more developmental cycles
and the B, C, and D sets even more, requiring a number
of years before all of the listening and reading tests
reach the standards of quality that the EEC desires.
Once the battery of set-form tests are fully
developed, however, there is no guarantee that they will
continue to reflect the current state of affairs of
incoming students at Ehime University. Indeed, the
common tests are currently being formed to achieve an
overall average score of75％ among1，800students from
six faculties. Due to future social and educational policy
changes and possible redistribution of Ehime University’s
student population among its six faculties, the immutable
disposition of the set-form approach is a distinct
disadvantage.
In addition, since the common listening and reading
tests presently account for30％ of students’ final grades
they can be considered as high-stakes examinations
where security is a fundamental concern. Whether
accidental or incidental, any public release of a common,
set-form test would render it useless to future
application and the energies mustered in developing it
would seem sorely wasted.
Endeavoring to overcome these developmental
obstacles and to add enhanced features to the listening
and reading common test program, members of the EEC
secured an Ehime university Good Practice grant in the
spring of 2015 to develop a test Item Bank（TIB）.
Making the transition from the set-form approach to the
The Advantages of Transitioning from a Set-Form
Testing Approach to Test Item Bank Methodology
Mark Duane STAFFORD
English Education Center, Ehime University
1）The terms test“question”and test“item”are used
interchangeably in this article.
61大学教育実践ジャーナル 第14号2016
TIB is a logical and necessary step in the evolution of
the EEC’s common test program. This paper describes
the advantages of transitioning from a set-form, CTT
approach to a TIB.
What Is A Test Item Bank ?
A test item bank is essentially a large pool of
questions that can be selectively extracted to form a test
unique to each administration. Once the questions are
piloted and improved under CTT, the entire bank of
questions is given to a large number of persons with
diverse abilities. A Rasch analysis is then performed
using specialized statistical software ― Winsteps3．81．0
（Linacre,2015）in our case. Items that are found not to
fit the measurement objective well may be revised, or
rejected from inclusion in the TIB（Bond and Fox,2007）.
Items assessed as having good fit are included in the
TIB and catalogued according to their difficulty relative
to the average person’s ability as determined by the
Rasch analysis.
Once the difficulty of each item is known, individual
questions（or blocks of questions as with the common
listening and reading tests）can be withdrawn from the
bank to form a test with a predictable and very accurate
overall average difficulty（mean score）.
Item Response Theory（IRT）was selected to be
employed in the design and analysis of the TIB because
it is based on establishing a model that specifies the
probability of observing each response option to an item
as a function of the target trait being measured by the
assessment, which is often a knowledge, skill, or ability.
In testing situations where items are scored as correct
or incorrect, IRT specifies the probability of a correct
response to an item as a function of ability. （The
University Of North Carolina At Greensboro,2015）
In addition, the one-parameter model of IRT, also
known as Rasch, was chosen over multiple-parameter
models because it ranks item difficulty the same for all
respondents independent of ability and person ability
independent of difficulty（Bond and Fox,2007）. There-
fore, it becomes relatively easy to form an objective
ranking of item difficulty with the information that Rasch
analysis provides and to use such information for the
TIB.
The Benefits Of A Test Item Bank
Although considerable effort is involved in writing
questions and conducting the Rasch analysis for the TIB,
the benefits of such labors are substantial. As
mentioned earlier, the problems with the set-form
approach can be solved and additional advantages can be
added to the listening and reading common test
programs through developing the TIB.
Time Savings
Significant time can be spared by creating the TIB
in comparison to going through the tedious
developmental cycles of the set-form approach. Once a
significant number of questions is generated, improved
through CTT, and a Rasch analysis is performed, misfits
can be excluded from the bank. Only items which fit
the measurement objective remain in the bank and their
difficulty is known and recorded. Although the abilities
of persons will change over time, the difficulty of the
items will not. Therefore, the TIB development process
only needs to be conducted once compared to the many
years required for the set-form approach.
Security
In addition to the time savings that developing a
TIB can offer, there is also a substantial increase in
security. As stated above, the public release of a set-
form common test renders it invalid for future
deployment. In addition, the information on set-form
tests can be transferred to future generations of
students by word of mouth. The TIB solves both of
these problems by allowing testing program supervisors
to withdraw different questions from the bank for each
administration. This essentially means that, given a
relatively high number of items in the bank, no question
will be used more than once during a certain number of
years. Each test will be unique, yet also equivalent
in difficulty to prior administrations, by using the
information provided by Rasch analysis.
Adaptability
The advantages in adaptability of a TIB in relation
to a battery of set-form tests are manifold.
First is the ability to use TIB questions in both
criterion-referenced and norm-referenced contexts. The
listening and reading common A, B, C, and D tests were
designed exclusively for a criterion-referenced situation
intended to measure whether students attained a certain
level of listening or reading proficiency. In doing so,
most of the set-form questions fall into a narrow range of
Mark Duane STAFFORD
62 大学教育実践ジャーナル 第14号2016
Table1
Table2
difficulty determined by the level of proficiency set by
the EEC’s curriculum and its testing committee. In
which case, a75％ average score was desired to ensure
that a majority of students pass the test（and
subsequently the course）. The75％ standard holds up
very well with the entire population of the freshman
class of nearly1，800（minus dropouts and students who
exempted from the course）.
However, when each faculty is taken into
consideration, a different picture emerges. Due to
contrasting admission standards regarding English
proficiencies, each faculty demonstrates a mean score
which is different from the others, varying from the
medical faculty’s 84．7％ to the agriculture faculty’s
71．92％（FB, classes of mixed faculties repeating the
course is even lower at60．73％）as can be viewed in
Table1containing data from the common reading test
administered during the 2014－2015 academic year.
While individual faculty mean scores are lower or higher
than the target 75％ , the weighting of student
populations among the faculties determaines the overall
average.
Although creating items within a narrow difficulty
range was advantageous and valid within a criterion-
referenced context, fixing most questions at a specific,
relatively low difficulty level does not provide us with
information about higher-ability students, not to mention
students in the lower percentiles. In other words, most
of the current test questions are too easy for higher level
students and too difficult for lower level ones.
Therefore, developing questions of a variety difficulties
for the TIB will add needed latitude to the listening and
reading common test program, while still allowing for a
highly predictable desired overall mean score of 75％.
Including some questions on both sides of the difficulty
spectrum will provide more information about“non-
average”students for curriculum development purposes
and allow the TIB to be used for norm-referenced
purposes such as placement within a level-streaming
curriculum, admission into advanced courses, or in lieu
of a standardized test such as the GTEC（Benesse
Corporation,2004）.
Second, a TIB will allow the common test program
to adapt to changing English proficiency levels of future
incoming students at Ehime University. Whereas the
set-form approach can be slowly adapted to increasing or
decreasing levels, the process is somewhat akin to a dog
chasing its own tail. The process of adapting the set-
form format is quite slow which may lead to never being
able to catch up with changing levels. The TIB, on the
other hand, allows for a quick response the variation in
levels by merely selecting items for the next
administration that constitute an easier or more difficult
test than the previous administration.
Fairness
Furthermore, greater fairness for students taking
the listening and reading common tests can be achieved
by transitioning from the set-form approach to a TIB.
Even though two set-form tests may be developed to
produce identical mean scores under CTT（an
accomplishment which remains incomplete within the
EEC）, test administrators assume that each question is
of equal difficulty in relation to all other questions on
both set-forms for all test-takers, which is highly
implausible. Although results of two“identical”tests
may share the same mean, how that mean is achieved
may be very different（See Table2）.
In this highly simplified example, the means of
Forms X and Y are the same, yet Form X is obviously
skewed in favor of higher ability students. Were the
two forms administered to different groups in
consecutive years（not to mention at the same time）
means observed for each test would likely be very
similar, but the two forms could not logically be
considered“identical”, which raises issues of the fairness
of employing set-forms under CTT practices.
TIB offers solutions to this dilemma. Since Rasch
analysis assigns a difficulty rating for items in relation to
person’s abilities, we can accurately predict how difficult
the collective items on a test will be for each ability level
The Advantages of Transitioning from a Set-Form Testing Approach to Test Item Bank Methodology
63大学教育実践ジャーナル 第14号2016
Table3
using what’s called Test Information Functions
（TIF）（Bond and Fox,2007, Yu,2010）. Moreover,
Rasch analysis software conveniently allows the
comparison, and hence, balancing of difficulty for two or
more alternate forms using TIFs. Clearly, using a TIB
and Rasch analysis would add greater fairness to the
listening and reading common tests whether
administered over a span of time or during a single
session.
Accuracy and Predictability
Using Rasch modeling with a TIB also offers much
greater accuracy and predictability over CTT practices.
In a pilot study to be reported later, Rasch analysis was
used to predict the target mean of .750 for the2014－2015
academic year common reading test. Items were
selected from both Reading A and B forms to constitute
a collective mean of0．745, according to Rasch modeling.
The resulting mean of the actual test was 0．751,
meaning that Rasch modeling was accurate in it’s
prediction of the actual mean to within 0．006 points.
When the mean that would have been predicted under
CTT, which was based on results from the alternate A
and B forms administration for the2012－2013 academic
year, is compared to the actual mean, a difference of
0．027 points can be observed. Where CTT methodol-
ogy would have predicted the mean with an error of
nearly three percent, Rasch modeling predicted the
actual mean to within about half a percentage point.
Clearly, the TIB and Rasch modeling approach holds
significant advantages over the set-form and CTT
approach. （see Table3）.
Drawbacks
While the advantages of a TIB are many, there are
also a few drawbacks associated with employing the
TIB. For one, Rasch analysis on all TIB items must be
performed again if any major changes or additions to the
system are made. In addition, unless multiple Rasch
analyses are performed, items of poor initial quality must
be thrown out of the TIB rather than revised and
improved. This is perhaps not the most efficient way to
utilize the substantial human resources required to build
the TIB.
Conclusion
In addition to offering accelerated test program
development, the TIB, Rasch methodology approach
brings greater security, adaptability, fairness, accuracy
and predictability compared to the set-form, CTT
approach. Although creating a TIB initially requires
substantial time and labor, the long-term benefits are
clearly worth such efforts. In light of the substantial
benefits that a TIB could bring compared to the set-form
approach, the EEC would indeed be amiss if it were to
pass on the opportunity to take a great evolutionary step
in developing its common test program.
Writing of additional passages and questions for the
listening and reading common test TIBs is currently
under way. Piloting of the new reading questions is
planned during the 2015－2016 academic fall semester,
and listening questions during the early 2016－2017
academic spring semester. TIB questions will then be
revised and improved under CTT procedures. Finally,
TIB items will be added to the regular common test
forms as pseudo questions which will not be calculated in
students’ final common test scores. Then the entire TIB
will be analyzed using Rasch methodology, item
difficulties will be determined, and the questions will be
catalogued according to their type and difficulty
estimates. If existing courses endure approaching
curriculum reform efforts, official use of the TIB will
begin during the2017－2018 academic year. Even if the
curriculum is drastically altered, the opportunity to
deploy the TIB in one form or another will hopefully
remain.
References
Benesse Corporation（2004）. GTEC, Global Test of English
Communication. Okayama, Japan : Benesse Corporation.
Bond, T. G. and Fox, C. M.（2007）. Applying The Rasch
Model : Fundamental Measurement In The Human
Sciences. New Jersey : Lawrence Erlbaum Associates, Inc.
Brown, J. D.（1996）. Testing in language programs. Upper
Saddle River, NJ : Prentice Hall.
Linacre, J. M.（2015）. WinstepsⓇ Rasch measurement com-
puter program. Beaverton, Oregon : Winsteps. com.
Yu, Chong Ho.（2010）. A simple guide to the item response
theory（IRT）and Rasch modeling. Downloaded from http:
//www.creative-wisdom.com/computer/sas/IRT.pdf
The University Of North Carolina At Greensboro.（2015）.
Retrieved from http://erm.uncg.edu/oaers/methodology-
resources/item-response-theory/
Mark Duane STAFFORD
64 大学教育実践ジャーナル 第14号2016


The Advantages of Transitioning from a Set-Form Testing Approach to Test Item Bank Methodology

中曽, 久雄

再婚禁止期間の合憲性(最高裁大法廷判決平成27年12月16日)

Not a member yet