Survey of student views towards questions about data related to their studies

Initial objective

The primary objective of the initial survey is to help us in the selection and further development of verification algorithms. This narrow objective will expand as the project progresses until it becomes a broad "user acceptance" test plan for the working system.

Study design

Since the population of our interest consists entirely of Internet users, the proper methods of distributing our survey are e-mail or internet survey. In order to collect data from the users, we will send them e-mail asking them to fill in questionnaire. Those e-mail will contain either link to the web-page with questionnaire or questionnaire itself in pdf form. Once when the users answer on the provided questions, they can submit answers to web-server or by e-mail. This methodology is time efficient (usually it takes 72 hours to get responses), no-cost involve, and with flexibility advantages. Using this methodology, we need to take into account response rate and e-mail "bounce backs".

Sample size calculation

Calculation of sample size (i.e. how many users are needed to be interviewed) refers to using statistical procedure (also known as power analysis) to justify the appropriate sample size for testing a given statistical hypothesis. In order to estimate sample size for our survey, a decision about sample size must be made based on necessary degree of precision. Three factors commonly used to calculate the sample size are: i) the confidence level, ii) the confidence interval, and iii) the percentage of the sample that will choose a given answer to a survey question. Table 1 shows how these factors determine the sample size for the population sizes in range of our interest (5.000, 10.000 and 15.000) using finite population correction (FPC) factor. The results assigned to the “undefined” population size do not include FPC factor. Common confidence levels in survey research include 90%, 95%, and 99%. The calculations are performed for the worse-case scenario where the percentage of the sample that will choose a given answer to a survey question is estimated as 50%. This is considered the most conservative estimate because it is associated with the largest sample size since the percentage of users that will choose a given answer to a survey question is unknown for the evaluation study.

Table 1 – Sample sizes for the chosen population sizes and different confidence levels and intervals


It is also necessary to account for those users who will not take part in the questionnaire. Consequently, the sample sizes given in Table 1 need to be increased by the percentage of these users. Assuming that this percentage is 10% and that the confidence interval is 5%, a minimum of 300 respondents would be required for analysis.

Considerations for the response rate and email "bounce backs"

Since we expect a response rate of 60%, we accounted for this by dividing the sample size (n=375) by .60 to get a new sample size of 625. Thus, 60% of 625 = 375. Our experience with email addresses obtained from a list has also taught us to expect approximately 15% of all email addresses to bounce back. Therefore, we have accounted for this by decreasing our sample expectations by another 15% [375/(.60 - .15) = 375/.45 = 833]. To sum, the final number of people we will email a survey to is 833 with the expectation of receiving 375 surveys back from this sample. To help achieve our desired number of completed surveys (375), we sent a pre-notification letter, a survey notification email, two follow-up emails, and conducted telephone prompting.

Subgroups in the target population

We anticipate that feedback is necessary from four sub-groups in target population:

• Active users who regularly use an authentication service, e.g. students.
• Administrative users who have operational responsibilities for an authentication service, e.g. librarians.
• System users who have installation and maintenance responsibilities for an authentication service, e.g. IT helpdesk.
• Policy users who have financial or managerial responsibilities for an authentication service, e.g. head of department.

Ideally, feedback should be proportional to the level of influence each sub-group exerts on the adoption and usage of authentication technology. The proportion would be refined as the research progresses. Therefore, it is necessary to establish quotas (i.e. sample sizes for sub-groups) to ensure that our sample accurately reflects relevant sub-groups in target population.

Questionnaire letter and form

The survey was conducted using both an online questionnaire and an Adobe Acrobat PDF questionnaire as shown below.

Survey results

Summary of results from students at two departments at the Faculty of Technology, University of Portsmouth.

How comfortable are students about answering questions relating to:


How confident would students be about correctly answering questions relating to:



Back to top of page.