solved Fundamentals of Test Construction; Psychometrics, Part I Test Construction: Format

Fundamentals of Test Construction; Psychometrics, Part I
Test Construction: Format Considerations
The lecture segment you just watched reviewed advantages and disadvantages of structured responses. Now consider constructed responses (e.g., essay-type questions, open-ended questions).  
What are two advantages of a constructed response?
What are two disadvantages of a constructed response?
If you were taking a test, what type of response—structured or constructed—would you prefer to complete, and why?
Test Construction: Item Wording

It is certainly not the case that I have never been bothered by feelings of anxiety upon entering a room full of strangers.
Why is this a poorly written item?

Try rewriting the item here to make it clearer.
Types of Variables; Scales of Measurement
Assignment of numbers to members on a baseball team
Intelligence test scores (e.g., WAIS-IV scores)
Place standings in a race
Minutes to complete a task

Chose from this options:
Interval 

Nominal 
Ratio 
Ordinal 

Transcript-
So we’re now going to talk about how tests are constructed, how they’re made. And the first element that we’re going to discuss are formatting considerations when creating a test.
So when we’re deciding how we’re going to format a test, it’s necessary for us to think about, what is the ultimate purpose of this test. What are we trying to get at by using the test. And how are we going to interpret individual scores. So an example might help to illustrate this point.
Dorfman and Hersen give an example of a test that’s designed to identify individuals who are in need of some kind of remedial intervention in a particular academic subject. Let’s just say that it’s math. And so we’re trying to identify individuals who are struggling in math.
On a test like this, having really difficult items– what we would call a high ceiling for the test– really isn’t as critical as having enough items on the lower end of the test– easier items– that can differentiate among different test-takers. Because, remember, the purpose of our measure is to identify individuals who are struggling in math. So this is an example of why we need to consider the purpose of the test in determining how we’re going to construct the test.
Now, one broad distinction that we can make when formatting a test is between multiple choice or other highly structured responses or items and what are known as constructed response items. So this is where the respondent literally constructs an answer, like might write an essay, or answer an open-ended prompt with a few sentences. These are different kinds of test items, and there are benefits and disadvantages of each type.
So if you think for a moment about a highly structured response, like a multiple choice response, one benefit of this kind of response is that it’s relatively easy to develop multiple choice type test items. And we can make tests fairly long by including these kinds of items in our measure.
A disadvantage is that we don’t get as much information from our respondent through a multiple choice item. We might sacrifice some unique information or rich information when we use that kind of format.
In contrast, a constructed response item allows an individual respondent to elaborate on their feelings or views on a particular item. It potentially gives us a lot of rich information.
A drawback is that these are difficult kinds of answers to evaluate. We have to come up with some kind of a scoring system or a rubric and train raters to be able to evaluate those answers, which is time-consuming and sometimes difficult.
Also, think through how difficult it would be to make a long test that’s composed of constructed response items. In other words, it might be fairly easy to make a 50-item multiple choice test, but think if you had a 50-item essay test. That would probably be too long for any respondents to be willing to complete.
So, again, there are benefits and drawbacks of each type of a response. And you can certainly combine these responses. You can have a test that includes both.