Research and Evaluation 1
1. What are some of the differences between primary and secondary data? Why should you use both primary and secondary data in research?
Data can be classified as primary or secondary. Primary data is data that has been collected from a first hand experience for example; data collected from interviews is primary. On the other hand, secondary data is data that has been collected and analyzed by someone else, for example, statistics in books and newspapers. One of the differences between primary and secondary data is that it is cheaper to collect secondary data as it involves going to the library, the internet and other sources to collect data. On the other hand, a researcher must go to the field in order to collect primary data. In addition, it is more time consuming to collect primary data as opposed to secondary data. It is important to use both primary and secondary data in research as this creates a firm groundwork for the basis of conclusions. In addition, a person can generate comparisons from what has been documented as opposed to what is on the ground.
2. How valid/reliable are the statistics presented on the evening news?
The reliability of data presented on the evening news depends on different factors like method of collection and analysis. However, statistics presented on the news is usually biased in order to create a certain impression. For example, if the weather is not conducive for travelling, statistics on the news will tend to show a higher percentage of accidents in order to influence people’s decisions. Therefore, statistics on the evening news may or may not be valid. It all depends on how data was collected and analyzed to come up with the presented conclusions.
3. What are the advantages and disadvantages of using surveys to conduct research? What is an example of a loaded question?
One of the major advantages of using surveys in research is that they are convenient for collection of data from large populations. In addition, the researcher has the liberty to choose what kind of data collection method to use in the research. If the questions are administered by the researcher, it becomes cheaper to collect the data. In addition, a researcher can administer questions via mail, internet or telephone. One disadvantage of using surveys in research is that the questions have to be highly generalized to fit the needs of the large population. In this view, data collected may be too vague for a researcher to draw a conclusion. The success of surveys highly depends on the respondents’ capabilities honesty thus may result in errors.
Example of loaded question: Have you stopped taking cocaine?
4. Compare and contrast probability sampling with non-probability sampling. Offer examples of each.
Sampling is categorized as either probability or non-probability sampling. Probability sampling involves giving the subjects an equal opportunity of being selected to participate in the research. This involves using random selection whereby there is no set criteria to be followed in selection. As opposed to probability sampling, non-probability sampling involves setting a criterion for which to select units. This means that some units have a higher chance of being chosen than others. An example of non-probability sampling is whereby a student conducting research on the opinions of students on a certain issue, chooses to interview students from his class because they are his friends or he can access them easily. On the other hand, in probability sampling, the student would collect all the names of the students in the university and randomly select students that would represent the whole university.
5. What are descriptive statistics? What is the relationship between an operational definition and descriptive statistics?
Descriptive statistics are statistics used to describe the main attributes of the data collected. In addition, descriptive statistics are used to give a general overview of what the data represents. Descriptive statistics give viable measurements depending on the data collected. On the other hand, operational definition refers to the explanation of data that is formed from visible traits. The relationship between descriptive statistics and operational definition is that, the operational definition defines terms and their meanings in relation to the research, and is used to help analyze the data using descriptive statistics.
6. Does all data have a mean, median, or mode? Why or why not? When is the mean the best measure of central tendency? When is the median the best measure of central tendency?
All data has a mean and a median and can be calculated when the data is numeric. However, not all data has a mode since sometimes all data may occur the same number of times. This means that the data has no mode. In addition, the mode can be found even in non-numeric data. The mean is the best measure of central tendency when the data being is continuous and the units are smaller in number. In addition, when there are no outliers or distorting units, the mean is the best measure. On the other hand, the median is the best measure of central tendency when there are outliers or extremes in the data. The median is used I this case to trim the distortion factors of any data.