Chapter I. Studying of question types for language skills

2.2 Grammatical analysis of qiestion types for testing
There are a variety of types of grammar tests, such as fill-in-the-blank, multiple choice, sentence diagramming, composition-based and authentic assessment. In general, grammar tests try to assess students’ use of language in its totality as well as various parts and rules of language. Grammar tests might focus on skills such as sentence structure, using and identifying parts of speech, punctuation, spelling and capitalization. They might also focus on complex and compound sentences and sentence corrections as well as writing correct sentences and paragraphs.
One of the types of grammar tests is fill-in-the-blank tests in which students are asked to provide the appropriate word. For example, they might be asked which form of a word — such as "their," "there" or "they’re" — a certain sentence requires and then be asked to fill in the blank. For this type of test, students might or might not be provided with a list of words from which to choose. Another type of grammar test that is commonly used is the multiple-choice format. For this format, students are asked a question and then given several choices from which to pick, with one of them being the correct choice. Students might be asked to identify which part of speech a certain word is and might be given choices. They might also be asked what measures to take to correct a sentence, and they might be given several choices. Sentence diagramming is another activity that students might be asked to do on grammar tests. To do this, they must make a graphic representation for the parts of speech in a sentence. Students draw lines and write the words from the sentence in a predetermined format to show how each word functions as a certain part of speech. Composition-based grammar tests are another way to check for students’ learning and mastery of grammar concepts. For these types of tests, students might be asked to write sentences, paragraphs or even a short essay in which they incorporate various rules and types of grammar elements correctly. On this kind of test, for example, students might be asked to take two sentences and combine them correctly. They also might be asked to take a paragraph that is written incorrectly and write it in a grammatically correct fashion.
Authentic assessment is another way to check for students' understanding of grammatical rules. For example, they might be asked to take a newspaper article and make corrections or improvements. The difference between this and other types of grammar tests is that it tests using grammar and language in their natural format and authentic contexts. Although this still is an academic activity, it is more related to real-life use of grammar rules than other types of grammar tests are. The purpose of this study is to develop and analyse a grammar test, focusing on validity, reliability and fairness. Effective test development requires a systematic, well-organized approach, thus ensuring the veracity of the proposed inferences from the test scores,. This systematic approach is outlined in the paper, expounding on how the test developed from the initial stage of identifying the construct to be measured, to the final stages of administering the test. This is followed by a detailed analysis of the test results which highlights possible threats to the test‘s validity, reliability and fairness. It also places the construct map under scrutiny, and questions whether the construct was successfully tested. The test was developed utilizing the four building blocks propounded by Wilson (2005). Each building block is explained in detail which gives insight into the process undertaken to write and analyse the test. Multiple-choice questions are a type of closed question, and involve a participant being asked to choose one or more responses to a question from a list of pre-determined options made by the researchers. Using this question type is a common feature of quantitative research because they are quick and easy to read, administer, and analyse. The structure of a multiple-choice question will often depend on the aim of the survey in which they are used. For questions used in tests where the purpose is to test the accuracy of the participant’s knowledge, the list of response options will consist of the (only) correct answer and a series of distractor options. In contrast, for questions used in research the list of response options will consist of a range of possible answers, but there is no right or wrong choice because researchers are interested in identifying trends in their population of interest. In research, multiple-choice questions are used to gauge concepts like satisfaction, agreement, confidence, or the likelihood of something occurring. Notwithstanding the aim, multiple-choice questions can be formatted in several different ways. Multiple choice questions are often a good choice for large-scale research because they are easier to administer and analyse, particularly for studies that are primarily based on number-based questionnaires or surveys. This is because participants only need to consider and select from one or more of the options provided to them, rather than having participants come up with a possible answer themselves. Providing the same set of responses is also helpful in creating a degree of consistency in your sample. However, with multiple-choice questions it’s important to keep in mind that it may be difficult to reduce the range of possible answers you include to a select handful, particularly when the range of answers or options is large. Further, because participants have to choose from answers that are predetermined by you as researchers, this may promote a degree of bias in participant responses because the options are what you expect the participant’s answers to be. Therefore, where practical, it’s always a good idea for multiple choice questions to have an extra option at the end providing participants with a space to self-identify their answer if it isn’t one of the options listed.
Matching questions are made up of two lists of related items that students must pair up by deciding which item in the first list corresponds to an item in the second list. They are appealing to many teachers because they provide a compact way to test a great deal of information in a short amount of time. However, creating effective matching questions requires some time and effort. Advantages of Using Matching Questions:
Matching questions have a number of advantages. As already stated, they are great at allowing teachers the ability to ask a number of questions in a short amount of time. In addition, these types of questions are quite useful for students with a low reading ability. According to Benson and Crocker in Educational and Psychological Measurement, students with low reading ability scored better and more consistently with matching questions than the other types of objective questions. They were found to be more reliable and valid. Thus, if a teacher has a number of students who have lower reading scores, they might want to consider including more matching questions on their assessments. Hints for Creating Effective Matching Questions: 1.The directions for a matching question need to be specific. Students should be told what they are matching, even if it seems obvious. They should also be told how they are to record their answer. Further, the directions need to clearly state whether an item will be used once or more than once. Here is an example of well-written matching directions: Directions: Write the letter of the American president on the line next to his description. Each president will be used only once. 2. Matching questions are made up of premises (left column) and responses (right column). More responses should be included than premises. For example, if you have four premises, you might want to include six responses. Limitations of Matching Questions. Even though there are a number of advantages to using matching questions, there are also a number of limitations that teachers must consider before including them in their assessments. 1.Matching questions can only measure factual material. Teachers cannot use these to have students apply the knowledge they have learned or analyze information. 2.They can only be used to assess homogenous knowledge. For example, a question based on matching elements with their atomic numbers would be acceptable. However, if a teacher wanted to include an atomic number question, a chemistry definition, a question about molecules, and one about states of matter, then a matching question would not work at all. 3. They are most easily applied at an elementary level. Matching questions work quite well when the information being tested is basic. However, as a course increases in complexity, it is often difficult to create effective matching questions. On a short-answer exam you are asked to provide a concise, yet thorough, written answer to a question, usually using complete sentences. These exams test your ability to integrate what you've learned in the lectures, readings, and discussions, and to apply that knowledge. Short Answer Questions do not have a generic structure. Questions may require answers such as complete the sentence, supply the missing word, short descriptive or qualitative answers, diagrams with explanations etc. The answer is usually short, from one word to a few lines. Often students may answer in bullet form. They usually take less than five minutes to read and answer, many take less than a minute. Short answer questions permit larger sampling of content. They tend towards greater objectivity in scoring. More reliable and valid than essay questions. Supply type test items are those which can be answered by a word, phrase, number or symbol. Supply type test items can be further divided into short answer test items and completion test items. Short-answer items can be answered by a word, phrase, or number. The two types of short-answer items-question and completion-also referred as fill-in-the blank, are essentially the same except for format. A student is given only two choices for an answer in this kind of question: True or False. The question content can include an image or html code. When feedback is enabled, the appropriate feedback message is shown to the student after the answer. For example, if the correct answer is "False", but they answer "True" (getting it wrong) then the "True" feedback is shown. Note that the True/False question type in a Lesson module is quite different. It is effectively the same as the Multiple Choice question type in a Lesson, but with only 2 responses.
True/False Questions Disadvantages: Care must be taken as to not make the questions too simplistic. It is hard to accurately gauge a learner’s actual understanding of the material judging by the results of a test composed of true/false questions. Learners have a 50% chance to guess the correct answer to every individual true/false question. A test containing nothing but true/false questions has to contain a significant number of them for the results of the test to be relevant. True/false questions are very handy when you need to check whether learners hold any misconceptions about the material or misunderstand it. Make sure that the questions cover every key point of the material, and the lack of understanding of specific areas of knowledge will be plain to see. Best Practices For Constructing Tests With True/False Questions The first rule for constructing tests that consist of nothing but true/false questions is that you need to have significantly more questions than, for example, a test constructed of multiple choice questions. It is recommended to use no less than 75 true/false questions in a test. In addition, you should keep the following guidelines in mind: State your questions as clearly and unambiguously as possible.

• 
  Keep your questions short and to the point. Never try to cover multiple ideas or notions with a single true/false question.
  
• 
  Every question must present itself as a statement that is unequivocally true or false, and that does not allow any exceptions.
  
• 
  Avoid using negative constructions in questions.
  
• 
  Common misconceptions make fine true/false questions (example: “The Great Wall of China is the only human-made object visible from the Moon: true or false?”). Use them to check the learners’ understanding of the material.
  
• 
  Construct your questions in such a way as to make incorrect answers appear more tempting to learners who do not know the correct answer.
  
• 
  Randomize the order in which questions based on true or false statements are presented.
  
• 
  Do not introduce ambiguity to your questions by using words such as “sometimes”, “the majority”, “most likely”, “some”, “usually not”, “best”, “worst”, etc.
  

In Conclusion. True/false questions is the most common question type, and it is widely used in testing. Such questions are the easiest to construct, but at the same time present the most potential pitfalls that must be avoided. Learners can spot patterns in the way questions are constructed and guess the correct answers based on the questions’ grammatical makeup and other indirect hints. Tests consisting exclusively of true/false questions can be successfully used as a learning trigger to engage the learners and presents topics for discussion. This is probably where this question type shines the most. In addition, it can also be successfully used in daily tests that can be employed to:

• 
  Motivate the learners to further study the material.
  
• 
  Help refresh the principal information delivered in earlier lessons or courses.
  
• 
  Serve as a basis for discussion regarding the points of contention in the material.