Public Opinion Research – Some research techniques for data collection

29 November 2016


In  modern  societies  opinion  research  respectively  data collection, of any kind, is very important for further development as well as for the definition and measuring of various issues of society such as: social-demographic developments, political elections, consumer habits, tourism, etc. A very important aspect in this connection is the usage of the adequate methodology, methods or techniques to collect the respective data. This article is to present an overview about the research methods of public opinion and their application in Kosova.

Key words: Methods, Research, Public Opinion Research in Kosova, Interviews, Questionnaire, quantitative researches, qualitative researches, Face-to-Face, CATI, CAWI.


To research, nowadays, the public opinion does not mean just to have knowledge in different research theories, to follow exclusively the research process, it requires above all an entire knowledge about the respective society in all its varieties and to use the appropriate research methods for it.

With regard to the historic aspect, public opinion research can be traced back from time immemorial, the history of data collection.

First of all, public opinion was usually related to political issues and is dated in 1824 in the United States of America where it was used for the collection of indicators of political personalities, respectively for the election of President Andrew

Jackson in Delaware and North Carolina.

In Kosova, for example, the opinion research, when comparing with other countries in the region3, is very new regarding  the  scientific  aspect,  commercial  need  and  other issues in the field of society. The reason for that, as already known,  is  the  country  status  of  Kosova,  lack  of  respective frames as well as of research infrastructure. With the increasing research awareness and emergence of problems, the opinion research in Kosova has become indispensible and continues to be an actual aspect.

“Opinion” is often compared in literature with other terms such as “thought” or “attitude” depending on the context of research and interpretation.

“In the sense of thought, it is explained to have a closer coherence to the context than the attitude and it is in most cases cognitive. In general, thoughts are evaluative and due to that fact they are considered as a specific part of research orientation which finally can be called as attitude”.4

When  regarding  it  from  the  scientific  point  of  view,  the public opinion research is a part of the quantitative and qualitative research methods; by using the following methods: CATI (computer assisted telephoning interview), CAPI (computer assisted personal interview), FTF (Face to Face), CAWI (computer assisted web interview),etc. a research can be realized.

In order to realize qualitative researches those methods are used  which  offer  more  detailed  results  or  data  and  are expressed in words or text form.

They  can  also  appear  in  television  spots,  various productions, and newspaper articles but their results cannot be generalized. Quantitative researches do not present detailed results; they are typically shown in numeric form and can be generalized.    In    many    scientific    works    the    following differentiations of researches are made: “the quantitative researchers argue that their data are real, rigorous, credible and scientific. The henchmen of the qualitative methods dispute that their data are sensitive, detailed and contextual”.5 Trochim and Donnelly (2006) explain later on the fact that both research methods are quite similar:”All quantitative data are based on qualitative estimations and all qualitative data can be described and manipulated in numbers.”

Nowadays,   the   opinion   research   is   used   by   different disciplines such as sociology, communication science, political sciences, psychology etc. In the field of sociology public opinion research is mostly used to describe the different social classes by  classifying  them  in  groups,  and  regarding  the  different family types etc. A good example for researches of the social issues is “Die Arbeitslosen von Marienthal”6, where, by means of quantitative   and   qualitative   methods   (questionnaires, interviews, statistical data etc.) the factors in the field of social live, economy and psychology are measured.

Public opinion research in communication sciences is different in its verbal or written context whereas the research in political sciences aims to reach prognoses usually before elections take place meaning estimations of the voters and political actualities.

When considering the opinion research in psychology, there are experimental methods used by aiming the measurement or corroboration of different hypotheses.

In other words, to make the long story about the debates short,7  we can generally say that the public opinion research deals with the exchange of opinions and attitudes of large society groups which have in general common characteristics such as, the registered voters and owners of smaller companies in a respective city.8



Research process – methods

Generally speaking, the research process in Kosova does not differ very much from the processes in other countries. The theories and stages can be similar or implementable but the research methods can differ because of the infrastructure or other social factors. However, there is no exact data, percentage about the method which is used in Kosova the most.

In order to explain easier the research methodology we can base on the research plan respectively on the research stages meaning data collection.

In  the  following graphic is  presented the  whole research process  starting  from  the  problem  identification or  research idea to the conclusion and publication of the data.9  This work puts emphasis on the empiric stage, which means decision on the ways of research, methods of data collection by taking into consideration the level of their application in Kosova. The determination of research methods always depends on the need or aim of the research project. An essential part of a research, before beginning with the application of the techniques for data collection, is the composition of research samples which is indeed, due to the undefined population figures in Kosova, difficult  to  realize;  thus  it  is  necessary  to  concentrate  on  a certain number of places of domicile of the respective target group aimed by the research project. In most cases, the researches are done through canvasses. These are the tools to get information but there is a diversity of methods for data collection.

Concerning the communication in canvass there are some oral  survey  methods  such  as  (interviews,  Face-to-Face, telephone surveys) as well as written methods such as (canvasses via post, internet etc.).

Before starting with the research and methods of canvass one should take into consideration also the form of the questionnaire by which all needed data are collected.

Stages of the topic/research issue


Compilation of the questionnaires

When compiling the questionnaires you should consider the content as well as the formulation of the questions. Schnell/ Hill/ Esser10 assert that there must be three initial questions:

  1. ƒ  What kind of information is required?
  2. ƒ  What kind of formal structure must contain the posed questions and answers?
  3. ƒ  What  kind  of  contextual  structure  must  contain  the  posed questions and answers?

The following questions can be asked by means of the questionnaire:


While   compiling   the   questionnaires  you   have   to   pay attention to the words and formulation of the sentences, therefore the questions must be as follows:


The number of research institutions is nowadays increasing steadily thereby also the number of different researches. Therefore one should be careful when compiling the questionnaires as all questions have a certain relation to the next question. Especially the beginning of the questionnaire is of high importance as it has to do with the engagement of the respondent. It is important to avoid many negative answers e.g. “I don’t know” or “it does not concern me” at the beginning in order not to give the impression that the topic does not concern the respondent at all.

The design of the questionnaire is also of high importance, therefore it is recommended to compile it as simple as possible, and the answers should be close to the questions and written within one page. With regard to the length of the questionnaire, there is no concrete model.



As there is no scientific regulation about testing the questionnaires,  it  is  necessary  to  pretest  them  before  the research starts. According to Schnell (1991) the questionnaires should be pretested as follows:


Oral canvass

There is no doubt, that the best way to do a research is the interview method. There are different ways, based on the standardization, to get in touch with the respondents such as (direct contact, by phone but also in writing) etc.

The Interview is a research method, a conversation between two persons, where one person is posing questions and the other one responds them, a conversation which cannot be compared with every day talks. This kind of conversation (asymmetric)  Question-Answer  is  essential  for  many researches, which is characterized by the aims of the researcher, who leads the data collection. The standardization is also of high importance. Standardization means same conditions for all probands which can be reached by the neutral attitude of the questioner and standardized interview. In the field of interview researches we have the standardized interviews; semi standardized ones and those without a structure.

ƒ  Standardized interview is a canvass method, where  the questions, answers as well as the listing of them are fixed. The aim of this method is to compare the data in order to get   thereby   a   comprehensive   result.   This    kind   of interview  is  done  in  an  asymmetric  way,   where  the questioner proceeds according to the stated  questions in the  questionnaire.  The  advantage  of  this  way  is  that through this structure you get the respective information very fast. Among others, a further aim of such interviews is to collect specified information in order to corroborate different hypotheses for comprehensive researches.

ƒ  Semi-standardized interview is a precast method, where

the questions are fixed but the questioner has the right to change the course of questions depending on the subject of conversation. By means of this method the respondent has the possibility to express more himself meaning, on the other hand, that the amount of information increases. In this connection, the interviewer has to see to it that all questions of the questionnaire are included.

ƒ  A non structured interview means that the interview  is not made based on the questionnaire but by conforming to the needs and situation of the respondent  to get thus the necessary information. The advantage of this method is  that  there  is  a  width  of  information  which  can  be specified in detail.

When regarding the aspect of qualitative researches, we see various kinds of interviews such as, narrative, focused and discursive interviews. Furthermore, there is also the expert, the explorative interview etc.

As to the appliance, this method is easy implementable in the society provided that information has been collected and the symbolic, simulative facilities are existent. But first of all, a salutatory letter is sent in order to explain the aim and methods of the canvass.

To the wide spread methods of canvass, we count also the personal ones, the so-called Face-to Face– Interviews, which are, nowadays, mostly used in Kosova. The received information from the respondent is recorded in the exposed questionnaire. This method is considered to be the most expensive one among the research methods. Depending on the research topic, it requires a high number of questioners which must be trained and cover widely the research area; as a result it requires more time to realize the canvass and high travel costs throughout the locations etc.12 Developed countries, for example, use also other methods like: CAPI (Computer assisted personal Interviewing) where through the computer (special computer for canvasses) data are collected.

Because of high costs (for the frequent use of telephone) and new technology, nowadays this method is replaced by the telephone interview also called CATI-it (Computer assisted Telephoning Interview). This method, however, limits the mobility to create a representative research as it requires certain telephone lines. There are two different ways to create the telephone interviews, the central and the local one. The central one means that the data are collected through call centers and the local one means that the questioner does the interview independently.

A   high   advantage   of   this   method   is   the   short   time availability of the data and low costs for the realization of the project. But with regard to the appliance of this method in Kosova, we have to take into consideration the number of respondents and disposability of the telephony. However, this method proves to be practical in case of refusal of the already known methods such as the Face-to Face one. This method is further simplifies by the so-called CATI-Software, which selects the numbers based on the randomization which must be included in a research. Another advantage of this software is the setting of dates, depending on the availability of the respondent during the research. In order to get a more representative result, at the beginning of the conversation some parameters are fixed such as a conversation with persons over 18, by asking which family member got lastly 18 years of age which would be than the person to be interviewed.


Written canvasses

Written canvasses by means of a questionnaire are classical methods of quantitative researches. Here, the respondent answers the questions based on a standardized questionnaire with many alternatives. The answer is given by marking the categories by a cross or underlining them, a frequently applied method is the Multiple Choice one. This method is preferable to measure and quantify the issues, scientific hypotheses and requires always certain knowledge about the issues which are studied respectively researched. By using this method you cannot discover any problem. Further conditions to maintain the research process is for example a database about the respondent data which simplifies the research and makes it more representative. The advantages of this method are e.g., that a large number of probands can be canvassed and exact results are reached but sometimes there are also small problems due to refusal of it by the respondent etc.

The  canvasses  via  post  are  also  a  part  of  this  written method. This method is usually used for long term projects, but the disadvantage of it is that you do not get the results quickly and the eventuality to receive a filled questionnaire back is smaller. Here, you have to think on sending also a stamped (prepaid) self-addressed envelope to receive timely a reply. Sometimes you have to mention the delivery date of the questionnaire. In Kosova, though, this method is used very rarely.

A further written method is also that via internet. By means of this method, however, you cannot reach an overall research result as the number of the persons visiting, sometimes unintended, the internet site is uncertain. The most popular method is that of sending the questionnaires via email and getting them back in the same way to the collecting point. This method  is   mostly  used   by   big   enterprises  for   economic purposes, evaluation of  their  employees etc.  When deciding that  the  data  collection  should  happen  this  way  than  you should pay attention to the visual introduction of the questionnaire and to give clear instructions how to fill the questionnaire.

In Kosova, for example, this method is used very often by different news agencies and media13, this method is  usually used to get public opinion but the method of electronic mail is used very rarely or not at all.



The authentic research methodology in our country is currently in the formation phase. Even if we are in the initial phase, we have to evaluate it positively by basing on the experience of international co operations with respective institutions which offer measurements and more detailed estimations of different research issues. Because of the current infrastructure in Kosova the most used method is the face-to-face one, even though it is more costly, but it must be replaced (at least in some cases) by another more efficient method or to be combined to a “mix- mode research”.

Translated by Dafina Vezaj



Friedrichs, Jürgen: Methoden empirischer Sozialforschung. 14. Auflage.

Opladen: Westdeutscher Verlag, 1990.

Lamnek,  Siegfrid:  Qualitative  Sozialforschung.  Bd.  2.  Methoden  und

Techniken. Beltz/PVU. Weinheim, 1993.

Jahoda, Marie/ Lazarsfeld, F. Paul/ Zeisel, Hans: Die Arbeitslosen von Marienthal– Ein soziographischer Versuch über die Wirkung langanduernder  Arbeitslosigkeit.  Suhrkamp  Verlag,  Frankfurt  a. M, 1975.

Oskamp,  Stuart/  Schulz,  P.  Wesley:  Attitudes  and  Opinions.  LEA

Lawrence  Erlbaum  Associates  Publishers,  III.  Edition.  Mahwah

New Jersey, USA 2005.

Schnell, Rainer/Bernhard, Hill Paul / Esser, Elke: Methoden der empirischen Sozialforschung. Band 7. Oldenburg Wissenschaftsverlag,


Trochim, William/ Donnelly, P. James: The research methods knowledge base. Atomic Dog, Cornell University, Sun- Buffalo USA, 2006.


Internet Sources


  1. visited on 03.03.2009.
  2. visited on 03.03.2009.
  3. visited on 03.03.2009.
  4. visited on 03.03.2009.