[MUSIC] In this session of week four, when we're looking at objective testing techniques, we're going to look at some alternatives to multiple choice questions. If all you can write is bad multiple choice questions, perhaps these might work for you. So, again we're looking at making test-like, formal teacher-centric test items in terms of format. Next week, we'll look at interactive and informal, student-centric type formats. A common alternative to a multiple choice is to the binary choice, in other words, you only get one choice. It's either true or it's false. It's right, or it's wrong. And on the screen, you can see the example, "Brass is an alloy of tin and copper, true or false." And the way it's highlighted, it must be false. I'm not a metallurgist, so I don't really know, I'm guessing too. Obviously, there's a high guessing factor when it's one out of two. Now, there will be some kids who will always be unlucky and guess wrong, but a lot of kids will guess right even if they didn't know. So, one of the alternatives we have is to ask for, if this is wrong, what's the correct answer? So, we're now forcing this student to make a judgement - true or false - and if it's false, provide a correct answer. Considerably more difficult both in terms of effort required because now they have to write an answer and also, the cognitive demand of actually thinking through, "Well, what is the right mix?" So, brass is an element of tin and copper appears to be false and the correct answer is that tin and copper make bronze. You could make this into a multiple choice or you could even say instead of having one true-false, what would happen if I had a set of three true-falses at the same time? And where instead of giving one mark for each one, I'll give one mark if you can get the set right. Now, this is a much more interesting approach to curriculum because it says these bits of knowledge are actually interconnected, and knowing one helps you know the others, but they're all related information. So, it gives us a nice interesting way of getting at related material in the curriculum. The other thing it does is to get three things right by chance means that the probability would be one in two times one in two times one in two, which is one in eight. So, it's actually technically harder than a standard multiple choice question. But the bonus is we get knowledge about multiple aspects. If you were teaching alloys in science perhaps, you have to - in the multiple binary - indicate whether each alloy is formed by the two metals given. So, you've got four statements here, each with two contributing elements and an alloy, and the student has to answer true or false. And in this case, to get them all right by chance would be one in sixteen - considerably harder than multiple choice. But because the information is interconnected, the probabilities are probably lower than one in sixteen because the information is not independent. Here's another one: remember our question about George Washington where we asked about which of one of these things is not true? Well, by a simple redesign of this question from being focusing on which one is not true, we've asked which three things are true and now the student has to identify three of these things which are true. One is false, which one is false, but the emphasis is now on the truth; not the falsehood. And it removes confusion, there is no worrying about what the not means. And although these might be unfamiliar, with a little bit of practice, students can go, this one's true, this is false, this is true, and they get one, mark for getting that set of knowledge. And as a bonus, as a curriculum teacher you get to know that this student knows a group of knowledge meaningful around a topic. Another technique is the matching list, and in the matching list we can use this where within a topic, there's a body of related material. For example, in language where we want students to know what new words in a new language mean and what their definitions are or what examples are of each word. In geography and science, where we have terms or even drawings and illustrations where they have to match parts to a list of possible parts, in mathematics where we have equations and definitions. Anywhere where there's a list of material that's related to itself and there's a possibility of getting it wrong, and where we can match a target with an option. And the key here, is that these target lists could be drawings, not just words. For example, here's a straightforward geography, a list of four countries with eight capital cities, and they have to match each capital to its right country. Now, you can imagine with the drawing lines, if a child draws a line and then changes their mind and draws a new line, this could get really messy to mark. So, it might be easier to leave a blank space beside each country in List A and have them write the answer in List B beside the answer, rather than drawing lines. The other thing I like about this quiz is there's two cities for each country, one of which is the capital, and it isn't always the largest city that's the capital. Here's another example we saw while we were working with the Malaysian exam syndicate, where they have figures of tools with no labels on the tools, and they have to match them to the function that the tool creates, rather than to the name of the tool and in each case, instead of having to write out the whole word, because of the way it's designed, the student only writes a letter in the right slot for each of the four tools. Notice in all of these examples so far, there's more options than there are command terms that we're looking for. There's only four tools. There's only four countries. But there's seven or eight choices. The reason for this is really simple. If the lists are equal in length, when you get to the last one, there's no thinking. There's no choosing. The answer's already selected. So, there's actually one less question than you intended. So, when a student gets to the last one, there must still be some choosing. so we always suggest that if you're going to do this, it must be uneven, that we prefer the command list that they have to match on the left and the option list on the right. It's a simple convention. And making sure that each thing is lettered and numbered, somehow coded so that it's easy for the student to show which option I've picked for each command option. Techniques like these simply move the world into more objectivity. We know there's a correct answer. We know there's only one correct answer for each list. But it treats material not as just independent pieces, instead this is related and integrated material. Think about the skeleton in science, human anatomy, you get a skeleton and a list of words and you have to match them. Matching techniques are a common part of education. It's an important skill in intelligence, and it's an important skill in most curriculum areas, and test questions like this give you a lot of powerful information quite quickly. In the next section, we are going to look at other techniques for testing objectively what students know and can do. [MUSIC]
