Powered by ProofFactor - Social Proof Notifications

Data Collection Methods

Oct 28, 2018 | 0 comments

Oct 28, 2018 | Essays | 0 comments

Rate this post

Data Collection Methods

Question 1

Survey

Advantages of survey

  • It has high representativeness
  • Low costs/cost effective
  • Convenient gathering of data
  • Good statistical significance
  • No or little observer subjectivity
  • Precise results
  • Easy to administer
  • Development takes less time compared to other methods of data collection
  • Can be administered remotely through telephone, email, mail, mobile devices and online
  • When administered remotely it prevents or reduces geographical dependence
  • Has capability of collecting data from large number of participants
  • Can collect broad range of data
  • The standardized surveys are usually free from nay types of errors

Disadvantages of survey

  • Inflexible design
  • Not ideal for issues that are controversial
  • Possibility of inappropriate questions
  • Respondents may provide answers that are not accurate
  • Respondents may be uncomfortable providing answers that portray them as unfavorable
  • Close ended questions of survey may have lower rate of validity
  • Data errors due to non-responses

Semi structured interviews

Advantages of semi structured interviews

  • Easy to analyze and fairly reliable
  • Fairly sensitive and flexible
  • Large amount of detail is generated
  • Questions can be designed in advance allowing the interviewer to be prepared for the interview
  • It allows the respondents to express their views in their own ways
  • it can provide comparable, reliable qualitative data
  • has specifically targeted audiences

Disadvantages of semi structured interviews

  • No guarantee of the participant’s honesty
  • Inference of Cause and effect cannot be provided
  • Interview of flexibility may reduce its reliability
  • It requires interviewing skills
  • Need data analysis skills
  • Resource intensive and time consuming
  • Confidentiality must be ensured

Question 2

Experimental designs

It is often difficult to generalize the results of experimental designs which are often criticized based on the size of the sample, the sampling design and the time within which the research was conducted. However, generalization of results is often possible is the following conditions are met:

  • The sample should be selected in such a manner that it is completely representative of the population under study. This means that researchers apply non-random sampling techniques in order to ensure that all aspects of behavior, physical characteristics and economic conditions of the population are represented.
  • The sample should also provide data so that it is possible for statistics to be applied in the entire population. Since behaviors possible change over time, the data needs to be spread over different time lines and therefore different eras. This is the only way that the results can be generalized. The statistics need to reflect the difference in time.

Question 3

Positivist versus interpretive epistemology

Positivist epistemology is based more on scientific research where facts are as they are. However, even for the major scientific subjects it has become obvious that research interpretation is more subjective as well as relative to the existing knowledge and set of behavior. Two experiments conducted in the same conditions, using similar resources and by the same researcher are more likely to yield differing results. In addition two researchers conducting the same research using the same conditions, again are most likely to get different results. This is the basis on interpretive epistemology, where knowledge is interpreted on the basis of the researcher.

Question 4

Survey design

Surveys are often much easier to design than other experimental methods. The following conditions are necessary to ensure a successful survey design:

  • Define objectives: the survey design comes after the objectives have been set. The entire research works backwards from the objectives. With the objectives, researchers can determine which kind of data they require and therefore design the research to fit into the expected data.

Pilot studies are vital for determining the validity and reliability of the research design. Researchers choose a similar population, with similar characteristics upon which the pilot study is conducted. The study allows researchers to determine validity of the research instruments and reliability of the same.

  • Biasness is one of the greatest concerns in the survey design. Survey researchers are prone to asking leading question, implementing research designs that are considered to be leading in terms of selection and analysis of the data. It is therefore important to note the concept of biasness and take steps to avoid the same.

Question 5

Benefits of comparative analysis

  • Comparative analysis allows the researcher to compare two aspects of research and find out the effects of each on the other. This is vital for scientific research but also the social research, where researchers can determine which aspects of nature, economy and other variables affect and can influence behavior of individuals.
  • Comparative analysis can also be used to compare variables and changes in research over time. Spatial and time analyses are often the most common and most sought after forms of comparative analysis. Research conducted at different times is an ideal way to increase understanding of subject matter.

Question 6

Confidence level

Confidence level is often defined as the uncertainty of the researcher with regard to the sampling technique and the choice of population under study. Since the sample is often used in making generalizations, it is often important to determine the certainty level of population parameters. The confidence level has the advantage of determining the best sampling technique which will result is a lower confidence interval. As such generalization becomes easier and more reliable.

Question 7

Ethics in participant research

The main ethical issues in participant observation as informed and voluntary consent. Researchers face a dilemma where informed consent may change the behavior of the population under research distorting the true data. On the other hand, lack of information could lead to legal problems where data has been gathered without consent. In addition, it is difficult for the researcher to gain voluntary consent without affecting or altering the behavior of the population under research. Should the subjects opt out of the research, the researcher must relinquish all data concerned with them. All respondents need to be protected from any ill effects that could arise from the research and this falls on the shoulders of the researcher.

When it comes to participant research, explaining the objectives of the research in detail may not be necessary. This is because there are higher chances of the respondents and participants altering their behavior in the presence of the researcher. The presence of the researcher in itself is likely to change and alter behavior. This is because people change when something new or a new element in their environment. Data may also not be as generalizable as other forms of research simply because it includes the subjectivity of the researcher who in himself could alter results and become emotionally involved.

Question 8

Validity and reliability

When it comes to research establishing validity and reliability s completely vital to ensure that one has usable data as per the objectives set for the research. However, there is often a confusion when it comes to which is most important. Validity speaks to the content of the research instruments and their ability to gather the expected data. On the other hand, reliability speaks of the consistency of the instrument to gather the same data over time. It is important to note that when validity has already been established, it follows that consistency comes automatically and vice versa. Reliability is a component of validity, therefore establishing validity concludes that reliability has indeed been covered. Researchers, who focus on validity and reliability separately, tend to find that they become confusing and the process is unending and therefore futile. However focusing on ensuring that research instruments are valid, is simple in itself and ensures consistency as well.

Question 9

Quasi experiments

Quasi experiments are much easier to generalize as well as administer and set up. They are most ideal in conditions where the random sampling techniques would not only be possible but also quite unethical. For researchers, it takes much shorter time therefore to set up the experiment, conduct it and indeed establish the results. The researcher can also choose which variables they need to include within the element. Unlike other forms of research, quasi experiments allow manipulation of all variables as the researcher wishes.

However it is often hard to determine and experiment the effects of confounding variables. These are variables that affect the research data but are not necessarily observable such as attitude, feelings and emotions. In addition, the lack of random sampling brings in questions of internal validity which becomes quite hard and in some cases impossible to establish within the research. The data is also considered much weaker as opposed to other forms of research whose data can be easily accepted in academia.

Question 10

Tips for semi structured interviews

  • Prepare a guide: it is important to ensure that the guide covers important aspects of information that are necessary for the research. Lack of a guide often leads to collection of unnecessary information, waste of time and sometimes lack of data to complete some objectives.
  • Follow up: the interview allows for an opportunity to ask follow up questions, which provide an excellent chance to enrich data within one sitting. In fact in a sitting, researchers can collect an excellent amount of data.
  • Make records: it is possible to actually miss out on information even when listening keenly. Where possible make record of the interview allowing the researcher to go back and confirm information.

 

Don`t copy text!