A collect data is considered a cornerstone
Generally speaking, a collect data is considered a cornerstone for the achieving of any research objectives. Yet, the researcher cannot complete his research or study without data. Researchers must be systematic in the collection of data. If data are collected untidily, it will be difficult to answer research questions in a decisive way.
Data Collection is nothing more than planning for and obtaining valuable information on key quality features produced by your process. However, simply collecting data does not ensure that we will get appropriate or particular sufficient data to tell us what is occurring in our process. It is simply how information is gathered.
In the collection of data we have to be systematic. If data are collected haphazardly, it will be difficult to answer our research questions in a conclusive way. There are various methods to collect data: Using available information , Observing ,Interviewing (face-to-face) ,Administering written questionnaires ,Focus group discussions ,Projective techniques mapping and scaling .Those techniques of data collection allow us to systematically collect information about our objectives of study and about the settings in which they happen.
According to Sekaran (2007), we can obtain data from primary or/and secondary source. He declared that primary source refers to information which obtained firsthand by researcher; in other hand secondary data refer to information gathered from source already existing.
As we know that, after data are collected through previous techniques (questionnaire, interview and observation), or through secondary source .it's necessary to be edited. Analysis can be viewed as the classification, the aggregation into ingredient parts, and the manipulation of data to get answers to the research question.
Technology has facilitated data collection in many ways. With respect to observation, there are several mechanical and electronic means to observe behaviour. There are devices that count or track specific actions, and individuals' click-through behaviour can be tracked. Surveys can be administered electronically, and unstructured to structured interviews can be conducted electronically as well. Individual responses can be identified, categorized and compiled instantaneously. (Hair et al,2007)
When conducting research, there are two types of sampling routes that can be followed: probability and non-probability sampling. Probability sampling is used when data is collected and statistical inference needs to be made about the data. In other hand Non-probability sampling, on the other hand, is used when there is no way in which the probability of inclusion of any one person or unit in the sample can be predetermined (Saunders, et al., 2007:207-208)
Smith and Albaum (2005) declared that generally, process of analyzing and making inferences from sample data can be viewed as a process of improvement that involves a number of separate and chronological steps that may be recognized as part of three broad stages:
Tabulation: identifying suitable categories for the information desired, sorting the data into them, making the primary counts of responses, and using summarizing measures to offer economy of explanation and so facilitate understanding.
Formulating additional hypotheses: using the inductions derived from the data concerning the related variables, their parameters, their differences, and their relationships to suggest working hypotheses not originally considered.
3. Making inferences: getting conclusions about the variables that are essential, their parameters, their differences, and the relationships among them. In the tabulation process, appropriate categories are defined for coding the information desired, making the initial counts of responses, and preparing a descriptive summary of the data.
The competent analysis of research-obtained data requires a blending of art and science, of intuition and informal insight, and of judgment and statistical treatment, combined with a thorough knowledge of the context of the problem being investigated.
As we know that , Data gathered from surveys, or input from several independent or networked locations via data capture, data entry, or data logging. Secondary data can be collected by documentary data such as written reports, organisational databases, journals and newspapers. Reviewing area based data enabled the better understanding of the research problem on a local level. Lastly, survey data such as government reports, journals and survey were used (Saunders, et al., 2007:249).Researchers, through measurement, describe phenomena that exist in the business world in terms of factors such as demographics, behaviour, attitudes, beliefs, lifestyles and expectations of consumers and/or organizations. To describe phenomena, researchers must have data. Once data are obtained, it is then analysed and becomes the basis for informed decision-making, which in turn helps to reduce the risk of making costly errors. (Hair et al, 2007 )
Primary data collection methods can be divided into two types - qualitative and quantitative. There are two broad approaches to qualitative data collection - observation and interviews. Observational data are collected by systematically recording observations of people, events, or objects. This data can be obtained by use of human, mechanical, or electronic observation. A primary advantage of observational data is its unobtrusive approach, meaning the respondent is unaware of his or her participation in a research project. This avoids interview
bias since no instructions, questions or interaction between the researcher and respondent are involved. A disadvantage of this approach is there is no opportunity to observe any unseen characteristics, such as attitudes or reasons for the observed behaviour. Ethnographic research and content analysis are two special forms of the observational approach. An interview is where the researcher "speaks" to the respondent directly and can be structured, semi-structured (e.g., focus groups), or unstructured (e.g., depth interview and projective techniques). Case studies are also a means of collecting information about a specific event or activity.
Quantitative data collection involves gathering numerical data using structured questionnaires or observation guides to collect primary data from individuals. Surveys are often used to collect quantitative data and they fall into two broad categories: self-completion surveys and interviewer administered. Self-completion methods include mail surveys and electronic surveys. Major problems with any kind of self-completed questionnaire are the loss of researcher control and the very low response rate. However, a large amount of quantitative data can be collected. Interviewer-administered methods involve direct contact with respondents through personal interviews, either face-to-face or via telephone. Interviews are particularly helpful in gathering data when dealing with complex and/or sensitive issues, and when open-ended questions are used to collect data. Interviews also enable the researcher to obtain feedback and to use visual aids if the interviews are face-to-face. Also interviews are flexible in where they can be conducted and researchers can increase participation rates.
According to international development research centre, there are various methods to collect data:
Using available information
Usually there is a large amount of data that has already been collected by others, although it may not necessarily have been analysed or published. Locating these sources and retrieving the information is a good starting point in any data collection effort.
This term refers to technique that involves systematically selecting, watching and recording behaviour and characteristics of living beings, objects or phenomena
An interview is a data-collection technique that involves oral questioning of respondents, either individually or as a group. Answers to the questions posed during an interview can be recorded by writing them down (either during the interview itself or immediately after the interview) or by tape-recording the responses, or by a combination of both. Interviews can also be conducted with varying degrees of flexibility. The two extremes, high and low degree of flexibility.
Questionnaire also referred to as self-administered questionnaire, it is a data collection tool in which written questions are presented that are to be answered by the respondents in written form. A written questionnaire can be administered in different ways, such as by:
Sending questionnaires by mail with clear instructions on how to answer the questions and asking for mailed responses;
Gathering all or part of the respondents in one place at one time, giving oral or written instructions, and letting the respondents fill out the questionnaires; or
Hand-delivering questionnaires to respondents and collecting them later.
Questionnaire can be either open-ended or closed (with pre-categorised answers).
Focus group discussions (FGD)
A focus group discussion allows a group of 8 - 12 informants to freely discuss a certain subject with the guidance of a facilitator or reporter
When a researcher uses projective techniques, he or she asks an informant to react to some kind of visual or verbal stimulus.
Mapping and scaling
Mapping is a valuable technique for visually displaying relationships and resources. Mapping a community is also very useful and often indispensable as a pre-stage to sampling .in other hand, Scaling is a technique that allows researchers through their respondents to categorise certain variables that they would not be able to rank themselves .Therefore, mapping and scaling may be used as participatory techniques in rapid appraisals or situation analyses.
As we know that the sampling is very important in any survey research because it helps to reduce the cost of this research, it's not usually possible to collect data from all of the people. It is mean the act, process, or technique of selecting a suitable sample, or a representative part of a population for the purpose of determining parameters or characteristics of the whole Population. According to Veal (2006)A sample is selected from the population , the use of the term population is clear when we deal with community of people but in social research the term also applies in other instances ; for example the visitors to a resort over the course of year constitute the population of resort visitors .in addition to this the term population can also applied to non -human phenomena for example if a study of the physical characteristics of Malaysian beaches found that were 10,000 beaches in all ,from which 100 were to be selected for study , then the 10,000 beaches can be referred to as the population of beaches and 100 selected for study would be the sample.
Sampling can be defined as the selection of persons or units from the population to take part in the study, should it be possible (Dillon et al., 1993:34; Robson, 2002:260). There are also various options when choosing to do probability sampling. Simple random sampling is the first option and also the most straightforward method of probability sampling. It consists of randomly selecting persons or units from a population and is used when the population is relatively small and all of its participants are known. Simple random sampling becomes difficult and even impossible when the population size is very large (Leedy & Ormond, 2001:214-215; Robson, 2002:261). The second method is that of stratified random sampling. Stratified random sampling involves arranging the population into similar groups or strata. Respondents are then chosen at random from each group. This method is useful as it ensures equal representation from all groups within the population, but the improved efficiency gained from using this method is lost when there is much variability within a certain group or stratum. Within stratified random sampling there is proportionate and disproportionate sampling. Proportionate stratified sampling merely refers to sampling where the sample number in each stratum reflects the population as a whole e.g. when there are 80 percent males and 20 percent within the population, four times more males should be included into the sample. Disproportionate sampling is the exact opposite of proportionate sampling and is used to stress a smaller group's opinions within an unequal weighting situation (Leedy & Ormond, 2001:215; Robson, 2002:262).
Furthermore, there is systematic sampling which involves choosing samples at random using a preset sequence. This method has some drawbacks, as the full population list is needed and as a sequence is being followed to select the sample, certain persons will automatically be excluded from the sample. This has lead to statistical doubts as to how random this method really is (Leedy & Ormond, 2001:216; Robson, 2002:261-262). Another method is that of cluster sampling. This method requires the population to be divided into various similar groups with equal amounts of heterogeneous persons or units. This is done when, for example, the population is spread over a very large area geographically. The groups are then called clusters and random clusters are chosen as the sample. This method is less accurate than stratified random sampling (Leedy & Ormond, 2001:214-215; Saunders, et al., 2007:223). The last method of probability sampling is that of multistage sampling. Multistage sampling is an extension of cluster sampling. It involves selecting samples from within samples. It enables the researcher to decide on the scope of the research and to tailor it to the resources available for the research (Saunders, et al., 2007:223; Robson, 2002:263). Non-probability sampling, on the other hand, is used when there is no way in which the probability of inclusion of any one person or unit in the sample can be predetermined (Saunders, et al., 2007:207-208; Dillon et al., 1993:229). There are four methods of non-probability sampling that can be followed. The first of these methods is quota sampling. Quota sampling involves choosing respondents in accordance with the proportions in which they are found in the population. It is most commonly used for large populations and has the advantage of being inexpensive and convenient (Saunders, et al., 2007:228; Robson, 2002:265; Leedy & Ormond, 2001:219). The second method is that of convenience sampling. Respondents are chosen on the basis of availability and convenience. Data obtained through this method is not always trustworthy and the drawing of valuable conclusions might not be possible by using this sampling method (Robson, 2002:265; Leedy & Ormond, 2001:219).
The third method is snowball sampling where the researcher chooses individuals of interest from the population and interviews them. After the interviews, the individuals are then asked to identify other possible cases of the population who might be of interest to the study. The identified respondents are then interviewed and used to identify other possible respondents and so forth. This method of sampling is mostly used when it is difficult to find useful respondents within the population (Saunders, et al., 2007:232; Robson, 2002:267). The fourth method of non-probability sampling available is purposive or judgemental sampling. Purposive sampling is where respondents are selected, based on the researcherâ€Ÿs judgement. The elements are selected because it is believed that they represent the population, are of interest and will assist in satisfying the needs of the research (Saunders, et al., 2007:230; Dillon et al., 1993:229; Robson, 2002:265). Homogeneous cases will be selected, enabling a more in-depth study. However, there will quite possibly be typical cases, which will enable the illustration of the subject matter and critical cases, from which valuable conclusions can be drawn and important factors highlighted (Saunders, et al., 2007:232).
Hair et al (2007) declared that Qualitative data can be analysed by three steps :- (1) data reduction, (2) data display, and (3) drawing and verifying conclusions. Data reduction involves selecting, simplifying and transforming the data to make it more manageable and understandable. The process requires choices about what should be emphasized, minimized and eliminated, and initial choices are guided by predetermined research questions, but the analyst continuously looks for new meanings and relationships. Data display goes beyond this step by organizing the information in a way that facilitates drawing conclusions. This process helps researchers organize information and view it in a way that enables them to identify linkages and develop explanations that relate their findings to existing theory. Higher order themes or patterns are likely to be extracted from the data during this step. The final step is drawing conclusions and verifying their accuracy through cross checking. Drawing conclusions involves deciding what the identified themes and patterns mean and how they help to answer the research questions. Qualitative researchers begin drawing conclusions when data collection begins, not when it is complete. Verification involves checking and re-checking the data to ensure the initial conclusions are realistic, supportable and valid.
In other hand, Data analysis in quantitative analysis involves a series of steps: 1. Review conceptual framework and relationships to be studied,2. Prepare data for analysis, 3. Determine if research involves descriptive analysis or hypothesis testing, and 4. Conduct analysis .
In quantitative data analyses, it is necessary to prepare data before analysing them .because after data has been collected and before it is analysed, the researcher must examine it to ensure its validity. Blank responses, referred to as missing data, must be dealt with in some way. If the questions were not pre-coded then a system must be developed so they can be input into a database. Editing involves inspecting the data for completeness and consistency, checking to see if respondents understood the question or followed a particular sequence they were supposed to in a branching question and elimination of questionnaires with a large proportion of missing data or for which screening questions indicate an inappropriate respondent. Missing data can impact the validity of the findings and therefore must be identified and the problems resolved either by eliminating respondents from the analysis or estimating the missing values by substituting the mean. Coding and data entry means assigning a number to a particular response so the answer can be entered into a database. Data transformation, such as reverse coding, collapsing or combining categories of a variable, creating new variables, or calculating the average summated score, is typically done to more easily understand the data or achieve some other research objective.
Provide overview of data collection and data analysis ,explain the methods for collecting data.
Focus on important problem of collecting data and provide solutions.
According to Hair et al (2007) the main issues that need to be considered in selecting a method of data collection for a survey of opinions about diversity in the workplace the important point to consider is that opinions about diversity in the workplace are probably considered a sensitive area for many people. Thus, some employees might not be willing to share their true opinions when other employees are present during the data collection, such as a focus group. Anonymity is most likely also a concern for participants.
The data collection process must be full revealed so you will need to think through the process on a step-by-step basis. It's a good idea to list each separate step in the data collection process as if you were going to flow chart it. Review your list to insure that it is as logical and complete as possible. Test your list by applying it to several representative cases. Whatever method used to collect data, respondents will probably feel the need to remain anonymous. Perhaps a self-completed mail survey might be more appropriate than an email survey in which the respondent can be tracked.