Systems Analysis -- FAQ

Systems Analysis FAQ
Questionnaire/Interview Assignment
Some FAQ's and their answers have been provided by former students.
If some advice seems inconsistent with what Dr. Sauter has said in class, please contact her for clarification.

Why are we creating questionnaires for this assignment?

The goal of this assignment is to provide a questionnaire that you can use to survey your users. While you have the opinion of your "customer or client" about what works well and poorly in the office, the users of the service may have very different perspectives about the issue. Part of your job is to determine if the users have the same perspective. If they do not, then it is your job to reconcile the multiple perspectives of the problem.

Do we really need to get people to fill these out?

Yes, you will really need to distribute this to your users. The results from this questionnaire should be summarized in your final report as part of your justification for your specifications. How you do that is up to your group. Remember, though, part of what I will consider (in grading) is the how representative is your sample.

Must we have already gotten people to complete the questionnaires when this assignment is due?

You do not need to have responses to the questionnaire when the assignment is due. That will come later.

What should be included on the questionnaire?

Focus your questionnaire! Otherwise the information you receive from the users will not help you.

What needs to be justified?

You need to justify each question on your instrument. This should be done in one of three ways. If your question doesn't fit into one of these three categories, it probably should be dropped.

First, your question could be directed at determining whether or not the user perceives there is a problem and, if so, what that problem might be. Since you have some ideas from talking with your client, use those ideas as the basis for these questions.

Second, your question could be directed at determining whether different segments in your population view the problem differently. These questions should be aimed at segmenting your sample for your analysis.

Third, your question can have a direct link to a possible solution. So, knowing the answer to a question will help you decide whether or not to implement a solution.

How long should the questionnaire be?

Keep your questionnaire short. It should be around 10 or so questions. This will improve the likelihood that someone will want to fill it out.

What kind of questions should we use?

Use as many closed-ended (multiple choice-type) questions as possible. It will improve response rate and make it easier to analyze.

Where else can we get advice about questionnaires?

Don't forget to read your textbook and the "Dummies" book for suggestions about questions and approaches.

Towards whom do I gear my questionnaire?

You should gear your questionnaire towards the person who is doing it. Don't gear them towards management views. Try to make them unbiased.

What kind of questions should I ask?

Use as many closed-ended (multiple choice-type) questions as possible. It will improve response rate and make it easier to analyze. It is also a good idea, if you do include any open-ended questions, to place them at the end of the survey. You should put questions that will help you determine what the problem is you are trying to solve. Putting random questions will not help you in determining what needs to be done to solve the problem. The questions need to be focused and geared specifically towards the problem.

Is web based or paper based survey better?

Either one will work, because you can do anonymous surveys on both. But you have to look at your audience. Students will find it hard to get on the web and fill it out. While staff use their Internet all day and would find it less of a hassle to fill it out online during a free moment.

Can we create a second version of our questionnaire if we realized from feedback from the first one that we weren't asking the questions we need too?

Yes if time allows. Maybe you just want to revise the original one based on feedback, or completely start from scratch because you have changed your focus. The original questionnaire that you turn in is the only one that will be graded as an assignment item. However, including a revised one in your final project could affect the project grade accordingly.

If designing a good questionnaire comes with practice and I have none how will I pass this assignment?

You can read pages 210-213 in your book. They are very informative on how to construct a good questionnaire. You may also model yours after examples I have given you.

Should our questionnaires be directed towards the client or the user of the system?

The users. With your background assignment you should have gotten any needed information for the client, so surveying them would be pointless. The point of the questionnaire is to help you narrow down your search for the solution to the problem.

If you do not have a focus, can we use this questionnaire to get a focus?

Yes. Keep in mind that the questionnaire must be at longest 10 questions. Here are some examples: Project A, Project B, Group D, Group F, Group G, Group H. (Note: These links are to the first page of the questionnaires associated with the online project examples. You will need to reference the actual report to see the remainder of the questionnaire and its analysis.)

Can you use your questionnaire to get your focus?

Yes!

Can your questionnaire have a place for comments by the user or does it have to be closed-ended questions?

The questionnaire can have a place for comments that can help you with your final analysis and can provide information as to what the problem is.

What if my group doesn't have a focus on the problem?

You can use your questionnaire to determine a focus, but you must be very careful to construct your questionnaire to not only tell you where the problem is, but to give you some insight as to why it is the problem. Keep the questions general enough to discover what the problem is, but focused enough so their responses don't identify environmental factors or other problems that are beyond your scope.

Can we have more than one questionnaire? And if so, do we need to do this assignment for both?

Yes you can have more than one questionnaire. And no, you do not need to do this assignment for both. You only need to write a justification for one and turn that in. The choice is up to you which one you turn in; however, the benefit of turning them in is the feedback I will give you. If you do have more than one questionnaire, and turn only one of them in, I will only be able to offer feedback o the one you turned in.

Should we use an open or closed question format?

You should use an open question format if you are looking find qualitative data from the user. Use a closed question format if you need quantitative data.

Should every student turn in an individual questionnaire or can this be done in our group?

This is a group assignment.

What are some hints to help us focus our questionnaire?

Consider what you want to accomplish and what you want the questionnaire to do.

How long should the questionnaire be?

Any questionnaire you want filled out without a reward should be no longer than a page, the text should be large enough to read comfortably. Don't make the words smaller to cram more on a page.

How do I choose questionnaire respondents?

At times there will be more people to survey than you an handle and you must therefore decide which set of people to send the questionnaire to or which questionnaire to send to which group of people. Whichever group of respondents you choose, it should be representative of all users. In general, you can achieve a representative sample by any one or any combination of these four methods: those convenient to sample, a random group, a purposeful group, and a stratified sample. Detailed explanations of these methods can be found in your textbook.

Is there any time that I should use an open-ended question?

The only time you should really use an open-ended question is at the end if you were to ask for any additional comments. It decreases the likelihood that someone will take the time to fill out the questionnaire if you fill it with a lot of open-ended questions.

How many questionnaires should be distributed to the users?

You want a good sample of the population to get true results to your questions. The more questionnaires you have received the truer the results will be. You want a good sample of the population.

To who will I distribute the questionnaire to?

It will determine on if you need information from the client or from users of the system (the students).

How well must the questionnaires be written?

Since your questionnaires are in written form, they must be extremely clear in meaning and logical in sequence. Remember a person is completing a questionnaire, he or she only has the written questions to interpret and answer. You are not there to clarify each question's meaning.

Do we have to have a focus for this assignment?

It is more helpful to have one so your questionnaire can get you results. You don't have to have a focus; just your questionnaire can be used to find one.

How can I know that people are answering honestly and that the answer I receive are valid?

There is not way to know that the answers are valid. This is one reason why to keep the questionnaire short. If the questionnaire is shorter, people will be more apt to complete it, and complete it honestly.

Can you use the questionnaire assignment to determine a focus for the project?

Yes the questionnaire can be used for this but be sure to use close ended quest5ions and try to keep them simple and short. Individuals have trouble finishing a questionnaire when it is long winded and boring.

How do I use the results of the questionnaire in our project?

These results will help you more accurately define the problem. You will also be able to determine if other people agree with what your group perceives as the problem.

How many questionnaires should be passed out?

As many as is necessary for you to be certain that you have an area to focus on.

What needs to be justified?

Design of the questionnaire can be divided in to three parts.

First, Determine the question to be asked: This step determine a key link needs to be established between the project aim we are about to perform and the individual question we are about to ask.

Second, Select the question type for each question and specify the wording: appropriate word and questions can be determined through a combined process of exploring the word and thinking creatively.

Third, Decide on the layout and sequence: We need to lay out the questions and answer choices attractive and neatly. We need to be consistent in aspects such as wording and try to standardize by using as few question as possible. We need to avoid using lots of lines, borders and boxes since these can make the page looks to dense and deter respondents.

Why is it important to have your questionnaire addressed toward more than one person??

By addressing the questionnaire toward more than one person, you allow the opportunity to receive a wide range of feedback. Many people have different views and from these responses you can collect data to better analyze for a possible solution to whatever your problem may be.

When including multiple choice questions in our questionnaire, how many check-box choices do we give our users?

Depending on your question, you'll want to have a range of choices. Instead of just a yes or no choice, be more specific with your question and give them a range of choices from lowest to highest.

Would it be easier to do a group session or individual intervies for the questionaire assignment?

I think it would be more efficient to do individual interviews because getting specific answers from one interview can enable me to make some of my questions more specific for the next interviewee. I also think some people prefer to reply to some questions in private.

How will we know our questions are worth asking?

If you can justify your question so that it helps answer the problem you are trying to solve.

While doing the questionnaire assignment, how do we choose who will complete the questionnaire?

You will choose who will complete the questionnaire by finding out who are the users of the system and how they interact with that system.

Should we be trying to "sell" our ideas to the people taking our questionnaire?

No. The questionnaire should not try to sell your ideas. Rather, the questionnaire should present your ideas and ask the questions in such a way that it informs the questionnaire taker, yet still allows them to answer in their own opinion. Discuss possible answer choices for your questions with your group, and avoid answers that direct the questionnaire taker to make a "decide right now" kind or choice.

There three types of users my group can interview, what advise can you give us to help narrow down whom to interview?

Choose the user that you feel can provide information pertaining to designing the specifications for the next assignment.

What kind of information are we supposed to get out of the people we question?

You are trying to get specifications for you system.

How can I receive good quantitative feedback?

Use simple questions with choices provided for the user. You don't want the user to have to do a lot of writing or else they may not fill it out. You may also use scales, such as a scale that goes from one to seven, to determine the user's opinion on a topic. These suggestions do not guarantee good quantitative data, but they will increase your chances.

When we pass out these questionnaires, and the results we get back show that there is no need for our project, do we get a lower grade on the whole project?

Are we going to need to use statistical analysis tools such as z-tests or t-tests on our questionnaire results?

You are not required to use statistical analysis tools to analyze your questionnaire results, but using them may help you to filter the information you get from them. The questionnaires themselves should be sufficient enough for you to gain a better understanding of what needs to be included on the prototype of your system.

What are the prototyping steps we should take to help our group complete the prototype assignment?

Step 1: Functional selection-range of functions never same as final software product.

Step 2: Construction-emphasis on intended evaluation disregarding data validity.

Step 3: Evaluation-need a test plan and evaluation resources and all relevant user groups.

Step 4: Further Use-throw away learning vehicle versus component of a larger system.

Do I focus on the people or thing that will use the system or the finished product or thing that comes out the other end.

Focus on the thing that will use or go through the system to make sure there is a need for it to even do so in the first place. Maybe the system shouldn't be made.

What type of prototyping model should we be using?

You should be using the experimental model of prototyping because all groups should have already clarified information requirements with the client which refers to the sub-model "exploratory."