I - Keeping your objectivity
When you decide to carry out a survey, objectivity is one of the keys which should not be lost. The french dictionary Larousse defines objectivity by specifying that it is the quality of whom: "can disregard its personal preferences." Those preferences are all the small things which make our ideas, our opinions, seemingly self-evident.
Those things can cause several undesirable consequences. For example, if you use, for a question, a choice of answers, it may be possible that your choice do not take into account the vision of your customers. This can happen when you ask how customers learned the existence of the store (or the organization). For the owner of a store, the moment when people learned its existence is very significant. For the customer ? He might even not remember when he saw your store for the first time. So, the answers will likely be evasive, even unusable.
To keep your objectivity, you must use some strategies. The goal is to try to be closer to the point of view of those you wish to interview. It might be difficult since your research still remains to be done. Then, you might not know, exactly, to what you should be closer. To solve this problem it is useful to ask yourself some questions like :
- What is the behavior of the targeted customer ?
- What are its usual activities ?
- What is the importance of your service (or product) in its everyday life ?
Finally, you must not forget that for several people, it is easy to overlook details that seems important from your own point of view. Often, someone will notices things only if a problem arises. Thus, the new decoration of your store can easily be ignored, even if you observed a real effect (like clients coming back much often). So be careful of how you analyze comments related to what seems obvious to you.
II- Looking at a problem
Many of those who want to do a survey, have in mind something that can be called : exploratory study. This can give an enormous quantity of information, often very useful. But, this approach can also complicate the life of those who tries such exercise. Here are some aspects to be kept in mind.
First, identify the goal.
Each time you wish to explore a problem, it means you be begin an adventure. As in any voyage, it is necessary to target a priority. It is this priority that will ensure you that everything will be coherent. Thus, it is important to give you only one priority. The identification of more than one priority is to be avoided! That can camouflage the "woolliness" of your project. In such case, going further can be the same as walking in a mined field . You can never know when everything will blow into your face, while you carry out the analysis.
In a second phase, organize the gathering of information.
An exploratory study is often attractive because of the great amount of information that it enables us to reveal. This advantage can however become a nightmare. Without being properly organized, the information gathered is, sometimes, barely usable. To organize the research, it will be useful :
- To be sure of the linkage between your priority and the information you wish to gather.
- To define in advance how information will be analyzed. It might be possible that your methodology is not adequate relatively to the information you will gather.
- To identify the kind of information to be gathered in order to be able to classify your data.
- To identify the form of information to be gathered in order to prevent difficulties. The way the information presents itself, can sometimes, cause problems when you try to do the classification prior to the analysis.
Do not forget that the main quality of an exploratory research is the access to a gold mine of information. It is you who have to plan how to use the asset resulting from your survey. That can make all the difference between a gold mine and quick sands where each effort will bury you deeper and deeper ...
.
III- Protection of privacy
To carry out a survey (or all other types of research), we often have to ask to people to give information related to personal matters. Usually, the confidentiality of all answers is part of the expectations of those who will answer. But, this concern should be an important issue, even when the requested information is trivial.
For all these reasons, the researcher must use a method of data processing that will prevent individuals from being identifiable. To simply say that we will not divulge any name is not sufficient. There is always some probabilities of an information leakage. In the worst cases, people with whom the researcher have a contract, might want to gain some "extras" by trying to makes us broke our rules of confidentiality.
The researcher must then protect himself and protect its information. To have concrete assurances of confidentiality, various strategies about the management of data can be taken.
When you do a survey, it is useful to break any link between the list of those who are part of our sample from those who will actually answer our questionnaire. The goal is to render impossible to find the name of any individual who filled such or such questionnaire. To use numbers on questionnaires is not sufficient. If there is a possibility to know that the questionnaire "X" was filled out by the person number 24, a simple glance at the list and the confidentiality is broken. For more security, the use of numbers on the questionnaires (for compilation) can be made after all the answers are gathered.
When you proceed with recorded interviews, we can use only a small quantity of tapes. Thus, once the grid of interview is filled and checked with the help of the recording, the elements likely to identify the respondent will disappear. The tapes can be re-used at regular intervals. That will ensure the success of the procedure of confidentiality. At the end, it will only remain to erase the tapes of the very last interviews to avoid any risk.
You should never underestimate the risks related to probabilities of information leakage. Only a well defined methodology can be a true insurance.