critical assessment of research methods used
...a or country. In times of economic growth, sources of finance may be more readily available than times of recession. This importance is also expressed by Helberg (2003) “An example of a problematic sample would be if the instrument were developed during an economic recession; it is reasonable to assume that people applying for jobs during a recession might be different as a group from those applying during a period of economic growth” (2003: 1). With a population of over 60,000, the sample size of 101 households is very small. In order for the results to have significance, the quantitative data must be accurate and reliable. The accuracy and reliability of the survey comes into question with such a small sample size. “If the sample size of the survey is too small it raises questions about the validity and accuracy of the survey” (Tufte 1983). The author makes no mention to any response rate, this raises further questions for the reader, the reader must assume that all 101 households in the sample responded to the survey, if this assumption is wrong, how many households responded to the survey? It is also important to be made aware of the way in which households were contacted, the author also fails to make any mention of this, describing the process simply as “a detailed survey was conducted of 101 households in Valladolid” (Biles 2004: 166). It is important to note if the survey was conducted over the phone, face to face or through the post. A ‘detailed’ set of questions may be too time consuming to complete and respondents may not complete the survey accurately. Green et al (2001) note that “two thirds of respondents had not filled in or had only given one line answers to the open ended questions at the end of the survey” Biles (2004: 166) also asserts that “Respondents were chosen randomly in several areas of the city”. There should be clarification as to whether these areas are representative of the city, are they suburban areas or city center slums? Although the author provides statistical data on the access to financial resources, no explanations are made of the data that is displayed. The survey found that 83% of households on a ‘very low’ income level used family and friends as a source of financial resources compared to 27% in ‘moderate to high income’ levels. However, no conclusions or explanations were made addressing these results. POSSIBLE IMPROVEMENTS TO RESEARCH DESIGN AND METHODS USED In relation to the sample size of the survey, a much larger figure should be used rather than randomly picking a number. A larger sample size would ensure that the survey is reliable and accurate. The results of the survey would consequently be more representative of the city in question and any result would not be so easily dismissed. The terms ‘low, moderate and high incomes’ should also be defined as the same word may mean different things to different people. To remedy this ambiguity, the author could re-classify the groupings into ‘below, average or above average’, once the average income is given, the reader will easily be able to relate any respondent to the particular grouping. A ‘detailed’ survey may also be inappropriate for a range of social classes. For example, households in poorer communities may not have the literary skills needed to fill out the survey which results in delays in responses or non responses. Unless the survey is collected, those in poorer communities would have to incur the extra expense of postage which many would be reluctant to do. Dilman et al (1999) note that, “a common observation among academics is that mixed-mode surveys are an increasingly popular way to maximise response while reducing costs”. The author should not solely rely on one mode of research (i.e. postal survey), instead telephone and face-to-face interviews should also have been used. This would result in a higher response rate particularly in the poorer communities as no costs are incurred. A brief description of the economic situation in Mexico should also be given, in times of recession where households would be experiencing financial hardships, this would result in an increase in loans taken out by households as a result of unemployment or general lack of income. This results in the survey loosing credibility and the results being distorted. A brief background to the areas in the city that were surveyed should also be given as it is important to establish if the areas concerned are high class suburban areas or working class inner city areas. The randomness of the survey also means that a higher proportion of low or high income households could have been surveyed....