Formulate a research question that reflects the research purpose, and that indicates the intended universe of analysis
The objective of the proposed research is to determine the proportion of the residents of Mamelodi living in poverty. This objective can be achieved by answering a research question that identifies the problem to be studied and the universe of analysis (Pretorius 32). Hence, the research question is: How many households are currently (2012) living in poverty in Mamelodi?
We will write a custom Essay on Social Study: Mamelodi Residents Living in Poverty specifically for you
301 certified writers online
Indicate whether or not the proposal writer should distinguish between the universe of analysis and the research population and between units of analysis and units of observation. Give reasons why he should distinguish between such entities (or not). Provide specifications that would, in your view, be appropriate
The proposal writer should distinguish between the universe of analysis and the research population. The universe of analysis illustrates the theoretically envisaged social entities that the research question focuses on (Pretorius 45). In this proposal, the household is the universe of analysis. It is important to specify the universe of analysis in order to determine the required theoretical scope of the generalizations that will be made using the collected data.
The research population, on the other hand, gives a clear definition of the boundaries, as well as the compositions of the social entities from which data will be collected (Pretorius 47). Since the universe of analysis, household, is abstract in nature, we cannot easily tell the specific group of households that the research question is referring to. For example, is it referring to households that consist of citizens of South Africa living in Mamelodi or every household living in Mamelodi? In order to eliminate this ambiguity, the researcher must distinguish between the universe of analysis and the research population.
The unit of analysis refers to the persons or the entities about which we must collect data in order to answer the research question (Pretorius 49). In this proposal, the unit of analysis is the households living in Mamelodi. However, the household is an abstract concept that does not clearly indicate the specific person (unit of observation) that will give data on household income. Thus, the proposal writer should distinguish between the unit of observation and the unit of analysis.
When distinguishing between the universe of analysis and the research population, as well as the unit of analysis and the unit of observation, the proposal writer should clearly specify the entities under consideration, their location, and the time period for which data will be collected.
Define the terms of quantitative interviews, qualitative interviews, standardized social enumerations, and participant observation
A quantitative interview is a method of data generation in which data are collected from respondents using interview schedules, which have predetermined questions and statements (Pretorius 57). The interviews typically involve an interactive question and answer session between the researcher and the respondent. A qualitative interview is a technique of data generation in which data are collected through unstructured or semi-structured questions.
This is done through a conversation between the researcher and the respondent (Pretorius 57). The researcher uses probing questions to solicit more information from the respondent. Standardized social enumerations refer to quantitative data that describe social events or occurrences. The data are often generated by private or public organizations and may include official statistics.
Standardization is achieved by collecting, processing, and reporting enumerations in the most appropriate data categories. Participant observation is a method of data generation in which data are collected through close and systematic observation of the life patterns of the respondents. The researcher collects data by actively observing and participating in the daily activities of the respondents or community of interest.
Which of the data-generation methods listed in 3 should the proposal writer use? Give reasons for your recommendation
The proposal writer should use quantitative interviews to generate the required data. This recommendation is based on the following reasons. To begin with, the Mamelodi area is relatively large in size. Thus, the researcher is likely to interview very many respondents. Consequently, conducting a face-to-face interview is likely to be very expensive and will also require a lot of time. In this case, the researcher can reach the respondents through telephone in situations where he/ she can not visit the interviewees. Since data on household income tend to be numerical in nature, the research will be quantitatively oriented.
Thus, the structured questions used in quantitative interviews will enable the researcher to collect data from many respondents on several poverty-related variables. Additionally, the structure will facilitate data analysis with the aid of computer software (Pretorius 76). In this proposal, social enumerations will not be applicable since the available statistical documents give national (aggregate) data on household income rather than data for specific areas such as Mamelodi.
In addition, such data might not focus on the specific type of households that the researcher is interested in. The data might also not be for the period under review. Participant observation might also not be applicable due to the metropolitan nature of Mamelodi. Concisely, the residents might not readily allow strangers to observe them. Furthermore, participant observation requires a lot of time, which might not be available.
Define the concepts of longitudinal research, cross-sectional research, and comparative research
Cross-sectional research is a study that utilizes data for a single time period (Pretorius 84). In this case, the collected data reveals the features of the units of analysis or respondents as they are during the time of data collection. The data is normally collected over a very short period of time so that the passage of time does not adversely affect the values of variables. Comparative research is a study that endeavors to determine the similarities and the differences between units of analysis (Pretorius 84).
A relatively small number of units or cases of analysis are normally considered in comparative research. This type of research is used to explain social phenomena by comparing the characteristics of the units of analysis. Longitudinal research is a type of study that enables us to follow changes in variables within a given period of time. Thus, it enables us to explain social phenomena by describing the changes in the characteristics of the units or cases over time. In general, longitudinal research enables researchers to study social changes that occur over time.
Which of the types of research mentioned in 5 is most likely to serve the research purpose? Give reasons for your choice
Cross-sectional research is likely to serve the purpose of the proposed study. As mentioned earlier, the main objective of the research is to find out the proportion of the residents of Mamelodi living in poverty. Since the research question refers to the current period (2012), the data on household incomes can be easily collected through cross-sectional interviews. The suitability of cross-sectional research is also based on the fact that we are interested in gathering data on household income characteristics as they are in 2012.
Get your first paper with 15% OFF
In this regard, a longitudinal research will not be applicable since we cannot collect data for any other time period to enable us to track any change. Since we are not interested in the changes in the incomes of households, a longitudinal research will not be necessary (Pretorius 85). In addition, significant changes in poverty levels or household incomes cannot easily take place within a single year. Given this time constraint, we cannot track any meaningful changes in poverty level using a longitudinal research. Similarly, a comparative research will not be suitable since we are not interested in comparing the poverty level or the household incomes of the residents of Mamelodi.
Considering your responses to the tasks and questions above, which of the following methods of data generation would be most suitable for the research purpose? Give reasons for your answer. The reasons must refer to both the relevant advantages and disadvantages (given the purpose) of each method
A structured and standard interview is the most suitable data generation method for the proposed study. This recommendation is justified by the following reasons. First, standardisation enables the researcher to ask the same question to all respondents (Pretorius 83). This makes it possible to compare the responses given by different interviewees. Second, standardisation facilitates collection of complete data from all respondents on the research topic.
Third, interviewer effects can be significantly reduced through standardised interviews. This is important since the proposed study might require collection of data by more than one interviewer due to the relatively large size of the geographical area to be covered. Fourth, the proposed study involves quantitative data that might require advanced statistical analysis. Data analysis can be simplified by generating data through standard and structured interviews. Finally, standard and structured interviews enable the researcher to ask several questions over a short period of time. This will enable the researcher to ask questions that sufficiently cover various aspects of poverty within a limited time. The disadvantage of standard and structured interviews is that they limit the interviewees’ response choices.
Hence, the meanings implied by the respondents can be distorted. This weakness can be reduced by reviewing the instruments in order to enhance their reliability and validity. Unlike standard and structured interviews, informal conversational interviews will enable the researcher to build the interview on observed characteristics of households (Pretorius 83). Moreover, it will enable the researcher to match the interview to specific households and their circumstances. Despite these strengths, informal conversational interviews will not be suitable since they complicate data organisation and analysis.
Furthermore, the collected data might not be comprehensive if some questions do not emerge during the interviews. This is likely to happen because some respondents are not likely to discuss the information about their household incomes with strangers.
Random sample of formal and informal housing units
A random sample of registered voters will be the most appropriate for the proposed study. In order to recruit a representative sample, the boundaries and the size of the population must be defined with accuracy. Given the difficulty in finding accurate information concerning the population size, samples are normally drawn from sample frames (Pretorius 93). Hence, a sample drawn from a list of registered voters is likely to be representative since the voters register contains an accurate list of people living in Mamelodi.
Additionally, it will enable the researcher to save time since the list of registered voters is readily available. Selecting a random sample of formal and informal housing units will not be suitable since it might not be possible to obtain a list of houses in Mamelodi or the number of people living in them. Similarly, obtaining a sample of heads of households can be difficult due to the lack of a sample frame. Besides, ‘head of a household’ is an abstract concept. For instance, we cannot tell whether the head of a household refers to the husband, wife or the first son in a household.
Pretorius, Louwrens. Thinking Methodologically: Research Methodology. Pretoria: University of South Africa, 2009. Print.