Online Surveys
What is an online survey?
We often recommend an online survey as a cost-effective method to reach a large number of residents. Island Global Research Panel Members are invited to complete the surveys we run. We reach a wide range of people, and have good reach amongst the older generations. Survey weights are routinely applied to ensure the findings are representative of the adult resident population by age and gender. If clients wish to undertake their own promotion in addition to ours then we can assist with this.
How we gather data?
We design a questionnaire based on the client’s brief. Our aim is to deliver useful and accurate insights based on high-quality data. The majority of the surveys we conduct are between 5 and 15 minutes long, with 5-10 minutes being optimum for maximum completion rates.
The surveys primarily contain closed questions with respondents required to select one or multiple options. In addition, we expect to include a small number of open text boxes.
The questionnaire would be set up and tested using our survey software. The professional software allows us to tailor the survey so that the questions asked of an individual may depend on answers to previous questions. Our online surveys can be completed on a computer, tablet or mobile phone.
An online survey would be launched and open to all island residents aged 16 and above. Eligibility criteria is determined following discussion with the client and an understanding of the required sub-group analysis, and screening criteria can be applied. We typically allow a 2-3 week period for data collection to allow as many people as possible to have their say.
We have over 11,000 people on our research panel across the Crown Dependencies including more than 4,000 from Jersey, 3,500 from the Bailiwick of Guernsey and 3,000 from the Isle of Man. We also extend participation to all residents via social media channels. Additionally, we often provide client specific survey links, can provide QR codes for printed material, as well as being able to design posters / adverts using In-Design.
Online data collection can be supplemented by targeted in-person, postal or telephone surveying if it appropriate to ensure minority groups are given the opportunity to participate.
We usually target a minimum sample size of 500 eligible adult residents per island. Typically, our surveys are completed by between 550 and 1200 eligible respondents. A sample size calculation which takes into account the adult population in Jersey, a 95% confidence interval and a maximum 5% margin of error indicates 383 respondents would be the minimum required sample size, assuming all adults who are resident in Jersey would be eligible. If the sample size were increased, the margin of error could be reduced further.
What is the output?
Once the online survey has closed, data are analysed in accordance with the research objectives. Segmentation and sub-group analysis can be important, depending on research objectives and how the findings will be used. We design the questionnaire with this in mind, often segmenting by attitudes and behaviours in addition to demographic indicators.
Survey weights are routinely applied to ensure the findings are representative of the adult resident population by age and gender. Alternatively, we can weight to ensure the sample is representative of households by composition and income. Survey weights adjust for any differences between the sample and the profile of the population. This is a standard statistical technique that allows us to present results that are representative of the local population.
Confidence intervals for proportions are calculated using the modified Wilson confidence limits. We check for statistically significant differences between sub-groups by identifying whether the upper and lower confidence interval for two results are overlapping or not. We usually include confidence intervals in an appendix of tables, for the overall population and for a range of sub-groups. This helps to quantify the statistical uncertainty that all sample surveys are subject to because they are based on a sample and not the entire population.
We offer different levels of reporting, with the option of basic, extended and comprehensive reports. The extended report is preferred by many Clients. It contains the overall results and selected results from the sub-group analysis – e.g. selected questions where sub-group analysis is of particular interest, or where there is a notable difference. High-level findings from thematic analysis of “free text” questions are included in the main report, and presented as a sorted list of comments as an appendix.
Our standard approach is to produce reports in Microsoft PowerPoint, in the IGR style and font, and provide the report to Clients as a PDF. However, we can usually accommodate a Client’s preferences if they want us to use a specific colour scheme, font or style. We can also provide data tables of results, online reports and infographics to assist in the communication of findings to stakeholders.
Market Research Society Standards
In all our work we adhere to the Market Research Society’s code of conduct. As part of the design process we ensure the questionnaire adheres to Market Research Society best practice guidelines, data quality is maximised and respondent drop-off is minimised. For example, we consider aspects such as flow, framing, question clarity and bias (wording & option set where applicable), respondent fatigue, appropriateness of question type, reliability of recall, and assessment of likely sample size for any sub-sets of questions. This also includes maintaining professional standards in obtaining and managing personal data for research purposes. Additionally, we comply with the General Data Protection Regulation (GDPR) and the UK Data Protection Act 2019.
