The validity of the measure is the extent to hich it measure the thing to you want to measure. Imagine, for eg, that you decide to measure a person''s general intelligence by placing a tape measure around that person''s skull at the level of his / her forehead, on the theory that larger skulls house larger brains produce higher intelligence. In this case tape measures is a valid measure of length but actually is it a valid measure intelligence? No, because measures of skull and brain size turned out to be unrelated to intelligence and it was later found that the largest brains on record belonged to retarded person.
Thus measures of intelligence based on skull or brain sizes turn out to be valid. b. Demonstrating Reliability and Accuracy: The reliability of a measure concerns its ability to produce similar results when repeated measurements are made under identical condition. The term accuracy means a measure that produces results that agree with a known standard. A particular measuring instrument can be inaccurate but reliable, but the converse is not true. It cannot be unreliable but true. For eg; a thermometer whose glass has slipped down in its bracket may read 21% each time it s immersed in water under ice water.
But as it gives same reading every time in the same conditions, it is reliable. If the thermometer reading does not agree with a known standard, as it is reliable, it cannot be accurate. c. Scale of measurement: Scales of measurement refers to the units in which a variable is measured, centimeters, seconds, 10 points and so on. Stevens identified four basic scales. From least to the most information they give, these are: Nominal scales Ordinal Scales Interval Scales Ration Scales. Q. What is sampling. Discuss in detail. A. Sampling is widely used in mass media research.
Surveys are the important part in research studies and form a distinct discipline in itself. Nationwide surveys require a Population and Sampling: Research areas can be investigated using human beings. When your research topic has to be verified y the investigation of human interest then you are require population under study. But it is impossible to study entire population, so it is easy to select a small sub-group from the larger should be able to give relevant information regarding your research topic. It has to be selected on the asis of importance and not availability.
Generalization: Your goal as a researcher is to apply the results of your experiment to the larger population from which the sample was drawn. You should not apply your results to those subjects only who participated in the study. Applying findings from a sample to a larger population is known as generalization. As the results of an experiment are to generalize the population, so we must be very careful while selecting the sample. The optimal procedure is to identify the population and then draw a random sample of ndividuals for participation in the experiment.
In random sampling, every person in the population must be eligible for population. They also should get an equal chance for being chosen. Random sampling is constructed by placing the name of every potential participant and then drawing out a specified number of names at one time. But in practice, this is not. Usually college students are used as they are easily available. The adverse effect of this is, it hampers the external validity or generality of experiment and the chances are that the opinion of non-college population may iffer in numbers of ways of college students.
Therefore, one should not limit themselves to college students only. Techniques of obtaining representative sampling is: a. Simple random sampling: Select a random number of individual from population. For eg; a golf ball. To combat problems here, select a large sample as it is more likely to represent all segments. b. Stratified sampling: It provides a way to obtain a representative sample. You begin by dividing the population into segments. You take people of a particular town, divide them in groups, then you separate random amples in equal size from each. Q. Discuss experimental research in detail.
A. The mass media are complicated phenomena. Apart from having a variety of media presenting information to the audiences, the audience use media in several ways. Some use them for information, and others use them either for entertainment or Just for time pass, creating a variety of research opportunities for investigation. A single research approach cannot be used because there are simply too many situations that needs investigation. Some situation require that subjects be studied under controlled conditions. Other situations require telephonic or personal interviews.
The experimental method is the oldest approach in mass media research and continues to provide a wealth of information for researchers and critics of the media. Advantages of Laboratory experiments: 1 . Evidence of causality: First, experiments help establish cause and effect. Although philosophers of science might argue whether we can ever really prove a cause - effect link among two variables, the experiment is undoubtedly the best social science method for establishing causality. In addition, the experimental method nvestigation. 2. Control: Control is another advantage of experimental method.
Researchers have control over the environment, the variables and the subjects. 3. Cost: In relative terms, the cost of an experiment can be low when compared to other research methods. An advertising researcher, for eg; can examine the impact of two different designs using an experimental design with as few as 40-50 subjects. A comparable test done in the field would be far more costly. 4. Replication: Finally, the experiment aids in replication. The conditions of the study are typically clearly pelled out in the description of an experiment making it easier for others to replicate. Disadvantages of Replication: 1 .
Artificiality: Perhaps the biggest problem with using the technique is the artificial nature of the experimental environment. The behaviour under investigation must be placed in circumstances that afford proper control. Unfortunately much behaviour of interest to mass media researchers is altered when studied out of its natural environment. 2. Experimenter bias: Experiments can have a problem with an experimenter bias. Rosenthal and Jacobson (1966) discovered that experimenters ho were told what findings were expected had results more in line with the research hypothesis than experimenters who were not told what to expect.
To counteract this problem, some researchers use the double blind technique, in which neither the subject nor the researchers know whether a given subject belongs to the control or the experimental group. Q. Write a long note on Readership Research. A. Readership research became important to management during 1960s and 1970s, as circulation rates in metropolitan areas began to level off and decline and editors nd publishers became concerned with holding the interests of readers.
They began more than ever to depend on surveys for the detailed audience information they needed to shape the content of a publication. Research into newspaper readership is composed primarily of five types of studies. The first called a reader profile, provides a demographic summary of the readers of a particular publication. For example, a profile of the audience of a travel-oriented magazine might disclose that the majority of the readers earn more than $40,000 a ear, are 25-30 years old, hold college degrees, possess six credit cards, and travel atlest three times a year.
This information can be used to focus the publication, prepare advertising programs and get increased subscription. A second type of newspaper study is the item-selection study, which is used to determine who reads specific parts of the paper. The readership of a particular item is usually measured by means of aided recall, whereby the interviewer shows a copy of the paper to the respondent to find out which stories the respondent remembers reading or looking through.
A variation in this technique is to have the interviewer preselect those items for which readership data are to be gathered and ask subjects if they remember only those items. Due to the expense involved in conducting personal interviews, some researchers now use phone interviews to collect issue of the paper is published. When the interviewer reaches a reader of the day''s paper, he or she asks the respondent to bring the copy of the paper to the phone. The interviewer who also had a copy of the paper, then goes over each page with the respondent, who identifies those particular items he or she has read.
Although this method saves money, it excludes from study those readers who do not happen to have a copy of the paper handy. Another money saving technique is to mail respondents a self-administered readership survey. Hvistendahl (1977) described two variations of this type of study. In the whole copy method, a sample of respondents receives an entire copy of the previous days paper in the mail, along with a set of instructions and a questionnaire. Q. Discuss method selection in research. A. Q. Discuss voluntary participation and informed consent.
No comments:
Post a Comment