An Instrument - Reliable and Validity - Assignment Example

Research Methods Define reliability and validity to include the ways of reporting each of them. how an instrument may be reliable and notvalid. Give examples. Reliability is the consistency of measurement or the degree to which an instrument measures the same way each time it is made use of under same conditions. In short it is the recurring of the measurement. A measurement is said to be reliable if a person's score on the same test given is similar on a number of occasions. Validity is the strength of the conclusions, inferences or propositions. According to Cook and Campbell (1979) validity is the best available approximation to the truth or falsity of a given inference, proposition or conclusion. There is a relationship between reliability and validity. A measurement may be reliable but need not be valid. Nevertheless a measurement has to be reliable before it can be valid. A tool should be reliable and should be suitable for measuring the intended object. In designing a research project it is absolutely necessary that the measurements are both reliable and valid. If it is not so the tools of measurement are basically useless for measuring what the researcher wanted to measure. 2. Almost daily, we hear about research that seems to contradict earlier findings. One day, for example, a particular food is good for you and the next it is not. What does this reflect about the scientific method What are the advantages and disadvantages of the scientific method It is very true that we find research to support or to conclusively oppose something. For any study that puts forward a recommendation there is another one that is apparently researched and documented and challenging the conclusions of the first. There is no convergence in the points of view of people. This does not invalidate the scientific method. There could may many possibilities for this divergence. Perhaps a researcher may not have used the valid tools for testing. The tools themselves may not have been reliable. The test conditions may not have been the same in all the cases. The theory put forward may have purposely stage-managed to establish a pre-determined outcome. The scientific method is closely associated with science. Modern human inquiry is pervaded by the scientific approach. It insists on systematic experimentation. Scientists have developed criteria and practices to reduce the individual and social bias on scientific findings. The basic approach is experiment, observation and inference. It approaches reality objectively. However, our five senses might even play tricks on us. That is why theories of Ptolemy were later replaced by the theories of Copernicus. Some times more efficient tools to observe reality will be developed which give more data, hitherto unavailable. The use of electron microscope has unraveled new insights into cell and modern cell biology is entirely different from what it was fifty years ago. 3. Suppose you wish to measure the effect of physical exercise on self-esteem. What type of experimental design would be most appropriate Describe the types of groups necessary for this design. How would you measure your variables Which threats to internal validity would you need to consider The available evidence indicates that exercise not only benefit by providing a healthier body but also enhancing self-esteem. However this has to be proved by empirical evidence. Visitors to a fitness center would be the ideal group to test the validity of this thesis. A group of twenty or thirty may ask to participate in the test. At least data from three stages are necessary to come to a reliable conclusion. Self- esteem measure before they start the fitness programme is the first step. Asking them to fill in a questionnaire specially designed to measure present level o f self-esteem can provide this. In the second stage the group should participate in fitness programme at scheduled times doing scheduled tasks. After the fitness program, which might last for 20 days or more it, is essential to measure the level self-esteem by asking the participants to fill in again the questionnaire. A comparison of the level of self-esteem before and after undergoing the fitness programme will validate the thesis. If the experiment is conducted in different fitness centers the reliability can be further attested by proving that results that are repeated in other places too. The possibility of an individual undoing an emotional problem which might have been got solved might cause the increase in self-esteem and may not be related to the fitness programme. The increase in self-esteem could have been due to fact that the fitness center provided social contacts that are needed for a person to feel a sense of self-respect. Another possibility is the presence of the fitness instructors can have an effect on the sense of self-esteem and devoid of the instructors the person may again relapse into lack of self-esteem even if he performs the exercise. 4. What closed-ended questions could you construct from each of the following open-ended questions a. What was your family's total income last year (Choose only one) 25000 to 35000 US Dollars 36000 to 45000 US Dollars 46000 to 55000 US Dollars Above 56000 US Dollars b. How do you feel about the space shuttle program (Choose only one) It is a good scientific study essential for humanity It is a good scientific study but not essential for humanity It has the value of sensation only It is unnecessary tampering with outer space c. How important is religion in your life (Choose only one) It is a great sources of strength in my life It is only an irrelevant social ritual in my life. It has a small role in my life I am not able to answer this question. d. What was your main reason for attending graduate school (Choose only one) For meeting people For cultural formation For acquiring skills for employment It is part of social behavior e. What do you feel is the biggest problem facing your community Excessive consumerism Excessive Individualism Racial and ethnic tensions Break down of the family life 5. A researcher designs a project to discover if women who are very career-oriented are less likely to have children than are women who are less career-oriented. He/she selects a random sample of women who are corporate employees and hold jobs at the vice-presidential or director level. Interviewing women he/she finds that their mean age is 45 and 35 percent are mothers, 55 percent are childless by choice, and 10 percent are childless involuntarily. She concludes that strong career orientations lead women to choose childlessness. Discuss some of the challenges to the validity of these conclusions. Think of a way this study could have been done. Specifically describe a sample and a study design that you think could be useful in doing research to answer this question. The conclusion of the researcher is not warranted by the method and tools used by him or her. First of all the 35 percent of the women who have climbed the top echelon of corporate world did so despite fulfilling their roles as mothers. So motherhood need not be a barrier to career growth. The fact that 55 percent are childless by choice need not indicate an exceptional pattern of behavior aimed at career growth. An assumption is warranted only in the light of significant deviation of from the existing patterns in the society to which the women belong. In many European, Australian and some sections of North American societies voluntary childlessness among large percent of women is rather the rule than exception. The test should investigate the motives for childlessness among women in a particular society then only the researcher can arrive at a reasonable conclusion. Since a test is valid only if the outcome is authenticated by repeated experiments, after remedying the above-mentioned errors it is essential to conduct a number of tests before arriving at a conclusion. To arrive at a valid conclusion, the researcher should undertake a survey of the percentage of women who are childless by their own choice. This part can be undertaken by survey handouts with open-ended questions on a sizable number of women in a society. Then through close-ended questions it is possible to find out the motives behind the voluntary childlessness in a society. This data has to be used as a point of reference to find out deviations in future studies. The study of the motives of the corporate women employees under consideration has to be conducted by using well-framed closed-ended questions. If the focus of study is to arrive at general conclusion about women the research should be undertaken across different geographical areas otherwise the conclusions might have validity only in a cultural context. 6. What is the role of statistics in social science research In describing a population, why is it advantageous to not only report measures of central tendency, but also to calculate measures of dispersion Be sure to distinguish between descriptive and inferential statistics. The application of statistics today is very broad and individuals, organizations and governments make use of it for arriving at informed decisions on variety of issues. It is widely used in the natural sciences, medicine, and business and in a big way in research related to social sciences. When statistics is used in social sciences it is called Social Statistics. It is the use of measurement systems to study human behavior in a social milieu. The key feature of modern social science research is the application of statistics and statistical analyses. Though statistical method is widely used in social sciences, nevertheless there has been lot of criticisms about the way in which it is used today to arrive at conclusions. Collecting data to analyze a phenomenon related to population studies usually produces large quantity of data. It is useful to summarize this large quantity of data in a single parameter. For research such summaries are necessary. In other words these summaries are measures of central tendency of the collected data. The most common of these measures are the mean, the median and the mode. However the measures of central tendency in population studies, for that matter in any study, does not provide clues regarding how the tendencies are dispersed with in the given set of data. Therefore dispersion is calculated by a variety of descriptive statistics like range, variance and standard deviation. The simplest measure of dispersion is called range. It is calculated by subtracting the smallest individual value of the data from the value of the largest. When presented along with the mean the statistic provides an idea of the variability. There are two approaches to the use of statistics in research. Based on this they are called descriptive statistics and inferential statistics. Descriptive statistics are used to describe the basic nature of the data in a research. They provide plain summaries about the sample and the measures. Along with basic graphic analysis, those form the foundation of all quantities analysis of data. Descriptive statistics are differentiated from inferential statistics. The descriptive statistics are merely describing what the data contains. But inferential statistics the research scholar arrives at conclusions that go far beyond the data. The result of a questionnaire plotted on a chart gives some pattern of responding to a question. The conclusion that we are able to derive from the responses of the population is the inferential statistics. 7. What are the key differences between probability and non-probability sampling designs Identify and briefly discuss the major types of probability sampling. There are different sampling methods available to the social science research scholar. One of the most rudimentary sampling methods is to work on probability and non-probability. In probability sampling related to population the investigator, selects randomly a certain number of respondents from a list. This is called random sampling. Second type is called systematic sampling. In this there is a definite plan for selecting the participant for the study. The second type of sampling is called non-probability sampling. In non-probability sampling the researcher decides the criteria for elements to be incorporated in the sample based on certain fixed parameters. The parameters could be acceptable conventional divisions like region, appearance, sex, age etc. Non- probability sampling also plays a major role in research related to social sciences. 8. Professor Hawkins receives funding from both a state agency and a national feminist organization to study the connection between marital power, spouses' financial resources, and marital violence. Deciding to interview both spouses from at least 100 married couples, she contacts potential participants by mail. Writing on university letterhead stationery, she tells each couple she would like to talk to them about their marriage, but does not specify the topics the interview will cover. a. What ethical principle(s) do you think was (were) violated in this research The fact that Professor Hawkins receives funding from state agency and a national feminist organization is cleverly hidden by the use of the letterhead of the University. This creates the impression that it is an independent study of a university. When funding is received from other bodies there is chance to manipulate the result to cater to the likes of the agencies providing funding. The person who is going to be interviewed has the right not to respond the interview if the subject of the interview is not acceptable to him or her. However, the participant in the survey is almost dragged to take part in a study without knowing on what he is going to be interviewed. It has violated his right to know what questions he will have to answer. b. Describe the probable reason(s) Hawkins violated the principle(s) you noted in your answer in Question 8a above. It was to camouflage the funding source that he was using the University stationary. The university stationary enjoys greater brand equity and hence perhaps enhances the good will of the researcher. c. Describe your reactions to the way this study was done. (Adler and Clark, 2003, 67) Adler and Clark (2003, 67) have explored the ethical concern centered on research. It is of paramount ethical importance that the participation is in voluntary. The consent should be also informed consent. These principles were not kept and hence the research of Professor Hawkins cannot be called ethical in the true sense of the word. Works cited Cook T. D., &Campbell, D. F. (1979). Quasi-experimentation: Design and analysis issues for field settings. Chicago: Rand McNally. Read More