item types in Reading
and their relevance to the assessment of student language services
Thomas Tinnefeld
Abstract
The present article investigates selected forms of assessment, their quality for testing purposes and their relevancy for evaluating students' performance in the area of reading comprehension, with the majority of findings generally being transferable to the field of listening comprehension. Potential advantages and disadvantages of the different types of assessment, such as summary, global questions, detailed questions, multiple-choice items, true-false statements, and answer matrix, are elaborated. Suggestions for their use in exams are given.
0th Introduction
The following will be about different exam and test tasks UNIcert ® context describe and to a view that assessment of student foreign language services, to assess. The aim of the paper is to make recommendations regarding the use or avoid certain Aufgabentpypen.
This paper is practical. He will help Klausurkonzeptoren and konzeptorinnen in preparing written parts of the exam, enabling them to focus in this way in an area where currently there is still very little guidance. Reading The we have chosen this for the following reasons as the subject of views:
The aspects to be discussed here in their scope in a way limitable which makes their treatment in a handbook article possible and appropriate. For wider issues rather the presentation form of a monograph would be useful. .
be presented in the following reflections are limited to a large extent not only on reading comprehension, but to convey to the listening comprehension and its testing. They have thus beyond itself.
Evaluation reading comprehension is the reason a particular problem because it is the mental processing of content and are sometimes put to very inadequate in objectifiable form. Example, this is different in the case of detection of linguistic errors in the texts students who will be identified in the rule to more or less objective way. For this reason, and Klausurkonzeptoren - konzeptorinnen criteria are given by the hand, it is possible to help the identification of a reliable evaluation results in a satisfactory and understandable form. This will be attempted here.
1. Item types
1.1 General considerations
to in our comments presented and by its design evaluated types of tasks are of central importance as they occur according to experience in the audit practice with high frequency. This does not, however, that other types - which are not covered in the following - there and in turn make sense. Thus, when lifted to the item types excluded here, but this place is a choice, but a well-founded. However, they are not a closed list. In some institutions be discussed here task types of the tested foreign language required - for example, as a summary of a French original in French - is in others the use of the language demands - that the summary of a French original in German (for German students). Both approaches have their advantages and disadvantages. However, both have their place, even if the use of mother tongue sometimes criticized. At those locations where it is necessary for understanding, addressing those fundamental distinction.
1.2. The item types in detail
The following are the relevant item types first described briefly, before going in the second chapter of their significance for the assessment of student language services. Specifically, these are:
- summary (résumé)
- Global issues
- detailed questions
- multiple choice tasks
- true-false statements
- response matrix
The Summary is the relatively free of most - and least controlled - in this case is eligible item types dar. Your goal is to present the original text in its main lines and him, centering on its core theme out. Here are essentially two different orientations possible:
- the chronological summary: This is played the original text in linear form. The sequence corresponds to the summary of the thematic sequence in the main text;
- systematic summary. This will form the achronologisch, ie it forms an imaginary center of the central text in the subject, grouped around which the various relevant factors. This form of the summary is generally regarded as the more sophisticated, the more significant. To what extent it is positive, the chronological summary is sanctioned, however negative, is the one the aspiration level of the teacher, leaving the other in the order of the examination requirements.
is undisputed, however, that the summary should not include more quantitative than a third of the original text. Summaries may include examples from the original text must, it is not.
Global issues serve large individual links in the main text to illuminate the substantive scope but ideally under which should be in the summary retrieved text segment. In practice, it is hardly possible to distinguish a series of global issues by an assembly, so that then the following relationship: Using a variety of global issues can be a given source text in a similar manner grasp how to use the resources of the summary. This fact is a fundamental problem of combining the two types of tasks: to recommend, therefore, is to require in the same exam either a summary or responding to global issues.
In detailed questions to receive a different situation: provide a useful addition to the summary of a part, and the response of global issues on the other dar. detailed questions to raise intentional from secondary or subordinated segments of text content. The objective of using this type of task is to determine whether the specimens have understood more than just the big picture of the source text. In answering the questions of detail they can prove whether they can set the main lines of the text to additional sensitive information in relation. Successfully answered detailed questions underpin the global understanding of a text and thus represent a reliable tool for a comprehensive review of reading comprehension can dar. uE go so far as to say that the testing of reading comprehension is not recommended using the summary or global issues without the support of detailed questions.
is required to create a summary or the answering of global or detailed questions builds up in the foreign language, it is essential to ensure that test segments are not in the main text by the students written off.
multiple choice questions differ from the previously discussed basic item types in that they of the specimens not require speech production. This object type exists in two forms:
- Only one of several possible answers given is correct and
- Several of the given answers are correct.
In the first case, it is not necessary to know the correct answer to solve a given item correctly: Candidates can either "guess" mark, so an alternative at random, the can then possibly be the right one, or they go ex negativo before and pass through the exclusion of all possible solutions to them seemingly unlikely the correct answer. This second option will not represent the intent by examining the form of solution, but it is - in contrast to the former - at least based language.
In the second case - the existence of several real solutions - which, however, in the field of language testing occurs due to systematic relationships much rarer, such evasion is not possible. For this reason, this form of multiple choice task in a foreign language examinations significantly upgraded be. Since in testing relevant usable utterances only in rare cases more than one correct solution exists or appear such statements will also have built strong, here is an example of a test-friendly application of this second form of multiple choice -problem are given. It might choose to introduce such an item as follows:
Pierre aime ______ _______ films.
a) les c) passion antes
b) d) passionnants
In an item like this alone, the combination of two possible answers, the correct solution (a and d) dar. Such a realization of this type of multiple-choice task is likely to be technically non-audit reliable than their traditionally carried out form. Challenging the teaching methods of testing, but also from a linguistic perspective it is all.
The True-False statements is to be played in virtually the same content in different shape or form a task type in the statements of the text. These statements must evaluate the candidate according to the categories of true and false. This test can occur in two main forms:
- in the form of pure check mark, so that by identifying a statement as 'true' or 'false' the problem is solved, or
- by the check mark in either category, and also - if 'false' must be checked - by playing the correct content in either the language or tested in the mother tongue of students.
This second form is to be regarded as challenging: It combines the largely passive element of the information identification with the active element in the formulation of correct textual information items.
It is also possible to perform here again in the discretion of the Klausurkonzeptoren, the offering of the correct items of information available in the tested language or in the mother tongue of students. As with the summary, both types of their advantages and disadvantages, but also its justification.
The response matrix refers to the working out of situations that occur in a given text: constellations between persons and things, between persons or things to each other with each other. Pairings that occur here would be, for example, what items belong to specific people or what arguments represent, what relationship they have at certain places and so on. These situations must be recorded in a matrix, by ticking or the granting of plus or minus sign for the validity or invalidity the individual relationships can happen. As the task type response matrix as a verbally passive element is classified.
second The assessment of student language services
This chapter is the assessment of student foreign language services, such as is possible by the previously mentioned types of tasks, in the center. The central question here is thus: In what form it will be with the different types of tasks, performance of students, they have to use the foreign language be valued?
The different task types are here singled out key issues and discussed with respect to the central question formulated.
2.1 Summary (Summary)
The first relevant element in the given context refers to whether the wording of the summary of the test samples clearly is. Since the clarity to be proportional to each other in the presentation of a summary and the success of the text understanding of the test can be regarded as that aspect of the sovereignty of text comprehension on the part of students. If the wording of the summary clearly, so that sovereignty is given with great certainty. Where the summary, however, a more or less disjointed presentation, the mental representation of the text in the candidate most likely will also be disconnected.
The question whether the central text message comes in the summary of test items reflected, takes off on whether they have understood the general thematic focus of the text. The negative response This question refers to a clearly failed this task. Here, the question raised be answered positively, however, this does not pass the reading comprehension portion of an entire test. Added to connect more then - the following - points of view.
For the monolingual approach - so that the foreign-language source text will also be reflected in the foreign language - another issue of relevance is: that if sections of the source text to be quoted verbatim. to request a summary is testing technology widespread. In such cases, where a Summary in German - or in the mother tongue of the students - no need is required, the question here.
The summary in the original language is the exact copy of the text by the students is to be avoided. Nevertheless, cited passages of the original text literally, the performance of students with regard to these passages is to question. Such acquisitions may have been made mainly for two reasons:
- The candidate formulations were detached from the text, not others to Available. This fact points to an inadequate command of written language, as the candidate Varationsmöglichkeiten lacking in expression;
- The candidate has not understood the passage in question and is for this reason, the literal adoption preference.
These were the reasons, if necessary, add more. All have in common is that they must be classified for the understanding of the text as problematic and therefore penalize negative in the assessment.
Another Criterion of evaluation will be the question of whether the information is in the main text of the specimens shown chronologically or systematically. Of course, a chronological presentation of this information is considered acceptable. However, it is less demanding than a systematic account. It can thus be considered as desirable: the less and the more systematic chronology can find in a text summary of the students, as the better their comprehension will be assessed. All the better, must be the evaluation thus.
A criterion that must not be underestimated, is in the question of whether the summary is designed so that it can be understood without knowledge of the source text. If the output text-independent course of a summary is not given, it must point naturally have a negative impact on the evaluation: Students will not be according to experience in their later working in a position to create text summaries, which need not be self-explanatory. For this reason, in this criterion is an important job-related questions can be seen.
also for the assessment of importance is the question of whether Selected students from the style of the summary is tight and concise, or whether "geschwafelt" in the summary. The text type Summary seeks naturally for a condensed style. Only when this is given, one can speak of a 'summary'. Summaries, which quantitatively greater than one third the length of the source texts are thus not worth more than such: The student is then an essential requirement for this type of text is not met. These relationships lead to penalize the vast length of summaries as a developmental disorder.
Another important criterion is the fact that all relevant information is to be found in the main text in the summary. A fundamental problem with this viewpoint is that different Klausurkonzeptoren may look quite different types of content of a text as relevant informational. Nevertheless, can the question raised at the best answer in such a way that for an "ideal" summary of the text in a total punk number is assigned. Each correctly reproduced information items will be rewarded by awarding a point. This type of evaluation is best for the specimens to understand and for the corrector at the comparatively most objective.
a question whose answer has significant problems, represents that whether there is any inaccurate expressed passages include a lack of comprehension. Then there can be no objective answer. But the reasons for this phenomenon are manifold. It urges, however, to conclude that lack of comprehension is compensated primarily by inaccurate expression. This conclusion is in accordance with the most realistic experience. It leads to value such passages negative.
A central question of evaluation of Summaries refers to the presence of misunderstandings on the part of the test: Is there such a view to the source text? Are negligible or that serious? In principle there are two possibilities for sanctions if the previously indicated point system is used:
- for a given misunderstanding can no points are granted or
- per misunderstanding will not be granted except a point, but it is also deducted a point.
For reasons of fairness in trials to be aimed at it is desirable on this question, the first possibility for the basis of the evaluation will allow.
If the source text contains examples, the question arises whether these are cited in the summary short and sweet. In fact, it can be assumed that examples have to text summaries, as a rule have no place. Still, to concede that a summary that contains no examples, can lose a substantial part of its information value: It lacks the illustrative character - an important aspect of understanding the text. Therefore, it may well be desirable in text summaries be to cite examples. This must be done clearly and concisely. If this condition is met, the outlined treatment of positive examples when the grade.
releases The points listed here as the main evaluation criteria for the type of text summary.
2.2. Global issues
For the task type Global questions are the following points are important: One important characteristic of this type of task is that the student recognizes the global nature of the question. He must understand that this is indeed a global issue - ie one that stands out to general content and trends in the main text. If the test this finding are clear, it must be inferred from the fact that he is aware of the thematic scope of the question and they can be classified in an adequate way in the whole text. This behavioral test is positive.
Central to the Assessment of global issues is the issue of whether - despite its general nature - are answered accurately and themes, so there is no deviation from the text and so too general statements can be avoided. This aspect, the orientation of the target item type dar. it is negative, there is a risk that the candidate shall not perish the statements of the text to answer the questions, but his knowledge of the world (see also Clapham (1998)). Such a behavior test for the assessment and therefore not desirable to sanction negative.
Another aspect that the position of global issues can be promoted, is that the specimens "escape" is not inquired in detail. It should be noted, therefore, whether they be for a evasion strategy. Such behavior of the samples may indicate that they have not grasped the central question addressed in the comprehensive text statement. This devotion to detail at those points where they can not be obtained, is to sanction negative.
As global questions on general developments and trends targeting is the answer to which is usually in a general diction to make orientation with a distant style. A concrete expression of the test and at the very Source text ajar style help, not to fulfill this objective. The assessment should be here accordingly.
As discussed in Section 2.1, the catchwords reference must be made on a text as well as on global issues as desirable: It is clear that the candidate the intratextual correspondences - the thematic links within the text - understood added. This is a positive aspect of performance.
The now to be addressed to the detailed questions are certainly factors to be considered as complementary to those here for the type Global questions have been addressed.
2.3. Detailed questions
One aspect that the quality of audit-related treatment of the item type influenced detailed questions by the student center, is that if the inquired detail by the student in Nutshell in its overall framework is provided. If this factor also has also the very character of the detailed question he is nevertheless regarded as a desideratum. He also picks up on the recognition intratextual correspondence, but here in contrast to the global question, they run in the hierarchy from top to bottom, go in the hierarchy from bottom to top. Both directions are mutually complementary and should be for the testing of reading comprehension also well used in this way.
If the inquired detail by the student named explicitly, this testing behavior makes it clear that he has put the question correctly with the text in relationship and provides a good overview of the text has, he understands it, the question and the content evoked by them adequately fit into the whole text. Although that aspect is nothing more than the present for the task type required to be provided learning goal, he is positive.
A student may act in an examination in a manner that suggests the impression that it evokes a different text detail than that on the aims of the question, and that detail has only indirect relevance to the question. Such testing behavior indicates that the student has trouble understanding, namely
- either terms of reference between the source and question or
- respect of such details, which aims at the question.
Assuming that the issue has been clearly stated - ie that no error on the part of the investigator is present - is more of this second option be considered. In this case, the power to sanction is negative.
also in the treatment of detailed questions is important to recognize it, whether such a question because of the text understanding or because of world knowledge of the test was answered. Although this problem is usually difficult to resolve because it is based on conjecture. In some cases it is possible to make such an attribution of the cause: then, if an example mentioned, this is not, or not in identical form in the text corresponds to emerging examples. Such testing behavior of a student points to a clear answer to the question based on his knowledge of the world as it is unusual to go on reading comprehension questions about the information in the main text. Such conduct should be sanctioned negatively, because the student has not complied with the rules of the game, this object type or recognize it may be that he has not covered the text with regard to this detail.
we were before - as highlighted in Chapter 1 was - to do with job types in which the student language production was demanded, this aspect of the following item type is not given.
2.4. multiple-choice tasks
the sake of completeness should be mentioned here that only the check mark a wrong answer alternative audit-related technical issues raised. Is the correct alternative has been checked, so this is aimed at the classic behavior of the students in multiple-choice tasks and represents the desired state. Relevant evaluation questions that arise in the first case, for example:
Has been checked an incorrect response alternative, but would have due to the formulation of the task can also be right? This question does not aim to be a problem on the part of students, but rather to a problem with the concept of the object itself occurs on such a case, an investigator with yourself be critical and to admit that the task formulated by him is inaccurate or unhappy. In such a case - and this is especially true if the check mark incorrect answer alternatives accumulates with students - may not be an examiner of this multiple-choice item be used towards assessment and sanction not negative. Such an item must be neutralized file.
have a number of candidates, including those where high performance is expected, as the same, the examiner did not expect response option is checked, this behavior most likely also a failure of the investigator and not to any fault of the students. Again, a design problem is given: the question and its associated response options are not clearly delineated. Also, such a case is not relevant for valuation. It should also be neutralized, or at best secondary importance included in the assessment. A total eradication is preferable.
Let us now turn to the penultimate task type: the True-False statements .
2.5. True-False statements
Also with this type of task is the correct check mark - either true or false - a classic performance of work and therefore not to discuss further. Also unmarked and thus not be considered further is the true substance of the claim for correction check mark by 'false'.
A problem is given if - at the check mark from 'false'-apart by the students made corrections to the content of the statement on other than the desired text detail. Such conduct is a reference test on the false understanding of the statement by the candidate dar. If the claim has been formulated precisely, and thus a wrong assignment of the source text objectively excluded, the test described behavior is to sanction negative.
Another way of testing student behavior is to have indeed checked = false, but make no correction of the claim. This behavior is either a matter of negligence of the students or the fact that they have not understood that is relevant for the correction of the statement text detail. You could also have come in answering the task because they Edit the content of the statement is linguistically impossible.
The one type of task that demands the students no linguistic expression is the response matrix.
2.6. Response matrix
under this item type in addition to the desirable situation - that the checkmarks are made correctly - essentially a problem identified point: are unsystematic, not based on the reading comprehension, Checkmarks been made? When this possibility first also happened to sound unlikely, it is found in the valuation practice yet. Such unsystematic check mark is present such as when the student produces conflicting assignments to the same facts or the same person. Such testing behavior clearly demonstrates a lack of comprehension and to sanction negative.
After this treatment made of the potential that is inherent in the mentioned types of tasks for assessing students' foreign language, with recommendations now to the separate task types.
third Recommendations regarding the use or avoidance of task types
As already implicitly an evaluation of the different types of tasks have been carried out is sufficient, at this point only a few hints.
recommend more or less full when it comes to assessing students' language performance in reading comprehension are the following types of tasks:
- Abstract (Summary)
- Global issues
- detailed questions
- True-False statements.
are less or not at all recommended.
- multiple choice tasks and
- response matrix.
This overview will be clear that these types of tasks are recommended to ask for a (certain) language production of students and include this potential in themselves - as for example in the True-False - is the case Statements , and that is when they are accompanied by a substantive correction in the case of 'false'-check mark. Types of tasks which require no linguistic activity of the students are, however, not recommended.
is for summaries and the use of global issues if they build in the tested language or to answer are made sure that no identical resumption of parts of the source text on the part of students to be produced, these two types of tasks to be assessed as positive warrant in an adequate and reliable manner, the testing of the text understanding in its broad outlines. The previously mentioned danger in the use of both types is that a series of global issues very often nothing more than a broken in the summary. Prüfungskonzeptoren must therefore ensure that when a summary is required, either no additional global questions are asked, or at most very few.
detailed issues are outstanding not only complementary to the task type Global issues, but also to the summary of . What these two types can not afford to make the detailed issues: the control of the main text as to secondary content aspects, thus regards content niches that without the consideration of this item type would not or barely detectable in language-productive way. Detailed questions are therefore generally unproblematic combine with global issues and also the type Summary . Since they do not affect the main lines of a text, but the content downstream factors represent detailed questions to those type of task with which the upper performance range of students can be tested for reading comprehension. The summary and the global issues are the basis of the text understanding, and hence the lower power ranges. Detailed questions represent the substantive details of the text - and thus to test the upper levels of the capacity of the specimens.
True-False statements are recommended on the condition that is required to offer the check mark from 'false' a substantive correction of the claim. If this correction by the examiner is not required, should be better but for this type of task. Will it be requested to provide True-False - is Statements an adequate alternative to multiple-choice tasks, as they whose most serious criticism - to compensate - not to require language production. One criticism is with them, however, that guessing the answer given is not completely excluded.
multiple choice tasks are to be rejected for two reasons:
- you When the potential is too high, guess the correct answer only. Because they allow themselves in the performance of four alternative answers is still a "hit rate" of 25% - which is clearly too high - they are a type of task extremely difficult.
- In multiple-choice tasks, it is hardly possible to ensure speech production in some form. They are completely passive and refuse on that ground alone.
multiple choice tasks are therefore not recommended unless it is used in Chapter 1 presented, but in the language field hardly apply task type with a cross where it is possible to have more than one solution. This variant of the multiple-choice tasks reduces the chances to guess the correct answer. Moreover, it approaches the point of view of speech production - even if no single word to be written - this indicates that the students in the check mark by more than one answer choice comparatively have to hire a deeper reflection than they have to do it, if only a single correct Reply to highlight alternative is.
is completely rejected as a task type of the response matrix. It is in its didactic value test is not powerful enough to see UNIcert ® entrance exams can. It may possibly be used as an exercise to "breathe" in the context of teaching, but not in the audit context, not as part of UNIcert ® .
4th Concluding remarks
The recommendations made in this article are greatly influenced by his own teaching experience and from our own examination activity. Foreign language instruction, think differently than it has been expressed here, may well do so in reasonable shape. Despite all the "subjectivity" of the statements made herein, however, appeals to the readers of the contribution, their own teaching and testing practices regarding the treatment here to check points and to compare with her. Even though they should not come to the conclusions that were drawn here, is proposed by the Comparison initiated reflection nevertheless constitute an important corrective to the teaching and examination practice of colleagues.
Literaturverzeichis
Alderson, Charles J. (2000): Assessing reading. Cambridge.
Eggensperger, K.-H. / J. Fischer (eds.) (1998): Handbook UNIcert ® . Bochum
Bausch, K.-R. et al. (Eds.) (1995) Manual language teaching . Third, revised and expanded edition. Tübingen / Basel
Clapham, Caroline (1998): The effect of language proficiency and background knowledge on EAP students' reading comprehension. In: Kunnan 1998: 141-168.
Clapham, Caroline and David Corson (ed.) (1997): Encyclopedia of language and education. Volume 7: Language testing and assessment. Dordrecht others.
Dlaska, A. Ch. Krekeler (2009): language tests. Performance appraisals Evaluate and improve foreign language teaching. Hohengehren: Schneider.
Groeben, Norbert / Hurrelmann Bettina (ed.) (2 2006): reading skills. Conditions, dimensions, functions. Weinheim / Munich: Juventa.
Kunnan, Anthony John (ed.) (1998): Validation in language assessment: selected papers from the 17th Language Testing Research Colloquium, Long Beach . Mahwah, New Jersey .
Nübold, P. (eds.) (2001): Foreign languages \u200b\u200bin higher education: What is university-specific language training? Bochum
Tinney, T. (1998): "French papers in UNIcert ® - a comparative view." In: Eggensperger / Fischer, 243-265
Tinney , T. (2001): "Technical Language exams in UNIcert ® context - with special consideration of the examination is" reading comprehension ". In: Nübold, 279-290
Tinney, T. (2002): Testing methodology. For the foundation of a new scientific discipline - the example of modern foreign languages. Aachen
Weir, CJ (1997): The testing of reading in a second language. In: Clapham / Corson 1997: 39-50.
See regarding a more detailed impression of existing UNIcert ® -exams and Tinney (1998).
See complementary to these linguistic skills and aspects of testing also Weir (1997), Alderson (2000 ) Tinney (2001) Groeben / Hurrelmann (2006) and Dlaska / Krekeler (2009: 78ff).
at this point to our monograph testing methodology. "(Tinney 2002), as raised here are discussed in greater detail in the questions.
l of science policy in inviolable spheres of-increases by many to be K rebsgeschwür our at every turn ailing and increasingly hollowed university system condemned by a few, however - and these are of course just to those with responsibility for the noble task of implementation - of these, however, he was chosen in his original goals exploited simply fundamentally perverted and tool of neoliberal Wirtschaftsd iktats . 
r at that statement was not as bad as one of the following extract from the Joint n Declaration of European Ministers of Education of 19th can be seen in June 1999 in Bolog na : 
