We use cookies to give you the best experience possible. By continuing we’ll assume you’re on board with our cookie policy

Statistical Package for the Social Sciences Essay

The author analysed data taken from a survey of attitudes to complementary therapies collected from people shopping in a town centre on a Wednesday afternoon. The survey took place in the high street outside a health food shop. For the purposes of this survey, complementary therapies were defined as: Treatments such as acupuncture, reflexology, aromatherapy, chiropractic, homeopathy, osteopathy, and herbal remedies.

The author entered data into a Statistical Package for the Social Sciences (SPSS) data file, examined the variables in this data, indicated the level of measurement and measure of central tendency that are appropriate for each variable. The author then generated suitable descriptive statistics for all the variables individually. In generating descriptive statistics, frequency distributions was used which according to Parahoo (1997) is the most basic analysis of quantitative data which involves counting the number of times a variable appears in data.

Frequency distribution is a statistical description of raw data in terms of the number or frequency of items characterised by each of a series or range of values of a continuous variable (Miller et al 2002). By using descriptive statistics, the author can organize, display, and describe data using tables, graphs, and other summary measures. Descriptive statistics will allow this author to describe a sample or population distribution in terms of its mean, median, modes, range, quartiles, variance, standard deviation, and outliers.

We will write a custom essay sample on Statistical Package for the Social Sciences specifically for you
for only $16.38 $13.9/page

Order now

According to Polit et al (2006), descriptive statistics are used to synthesize and describe data, averages and percentages are examples of descriptive statistics. In this survey, the sampling strategy used was quota, which is non-random. According to Bowling (2002), quota sampling is a method favoured by market researchers for its convenience and speed of sample recruitment. The interpretation of this survey data are that the frequency table show that 40 people took part in this survey and this included males and females. Frequency table indicate the sex of respondents as 26 (65%) female, and 14 (35%) male.

Frequency tables are used to record the number and percentage of responses associated with each response category of a variable. They are a commonly-used tool in univariate analysis and can be used to report on nominal, ordinal and interval variables. The pie chart also illustrates that 65% females responded, and 35% male responded. The level of measurement used for variable name “sex ” was nominal. The level of measurement refers to the relationship among the values that are assigned to the attributes for a variable (Trochim and Donnelly 2006).

Bowling and Ebrahim (2005) state that nominal value is a level of measurement where no ordering of cases is implied. The measure of central tendency is the mode, which is the most frequent value, (Parahoo 2006). The mode for the sex of respondents is female. The frequency table shows that the category of variable with the highest number of cases is Female with 26 cases against man with 14 cases. The descriptive statistics above appears to show that more females compared to males use complimentary therapies. Age of respondents, the table shows that 12 or 30% respondents aged 65+ participated in the survey.

The age band 45-54 had the least number of participants as compared to other age bands. The frequency tables shows that only 3 or 7. 5% of respondents were aged between 45-54. The level of measurement for variable name “age” is ordinal. Ordinal was used because the age of respondents on the data table was presented in age bands . According to Brown and Saunders (2008) ordinal is where the relative position of each case within the data is known, giving a definite order and indicating where one case is ranked relative to another. The mode for “age of respondents” is 65+.

The bar chart illustrates that respondents within the age band of 25-34 had the second number of participants followed by bands 55-64, 15-24 and 35-44 respectively. The findings from these descriptive statistics suggest that people above the age of 65 are more likely to visit a health food shop and that a small number of people between the age of 45-54 visit the health food shop. Views of respondents; respondents were interviewed to ascertain there views on whether that complementary therapies works. The frequency table indicate that a large number of people are of the view that complementary therapies are most likely to work.

The interpretation is also presented graphically and shows the category “mostly” on the X axis with the longest bar. The graphic is a bar chart with the categories at the bottom, the X axis, and the frequency scale at the left, the Y axis. The variable label “view of respondents” is displayed at the top of the chart. From the frequency distribution, 50% of the respondents are of the view that it is mostly that complementary therapies will work, 15% gave their view as “sometimes” 7. 5% as ” don’t know,” 10% as “hardly ever” and 5% as “never,” when they were asked if they think complementary therapies work.

The cumulative percent of those who were of the view that complementary therapies work is 12. 5%. The measure of central tendency is the median or mode which this author has defined in the above text. The median value of a set of data is the central observation, following the values being ranked in order of size. In other words half of the data will have a value less than the median, and the other half of the data will have a value greater than the median (University of Shefield ).

The frequency table for “answer of the respondents” shows that two responses were equally frequent. 0% out of 40 participants who were interviewed strongly agreed and 10% out of 40 participants agreed. The pie chart shows that 42. 5% disagree whilst 15% strongly disagreed and 22. 5% do not know. Pie charts are used to show the proportion of a variable in each category. The total for the data is presented by the area of the circle, and each category’s share of the total is represented by the area of a segment of that circle (Brown and Saunders 2008). In this case the categories referred to are; strongly agree, agree, don’t know, disagree and strongly disagree.

The cumulative percent column concerts the valid percentages into a running total summing to 100%. In the frequency table with variable label “answer of respondents,” for example, the cumulative percent shows that 77. 5% of respondents strongly agreed that complementary therapies only because the clients think they will. Responds of participants; the frequency table interprets that 27 respondents gave their answers as “yes” when they were asked if they have ever received any form of complementary therapy. 13 (32. 5%) answered “no” to the same question. The frequency table illustrates that 40 respondents were interviewed and none is missing.

From these interpretations it appears that more participants have received some form of complementary therapy. The limitations of this data are that the type of probability sample is accidental sampling. Parahoo -(2006, p267) states, that in accidental sampling only those available have a chance to be selected. In view of the survey it was a Wednesday afternoon and only those shoppers who were available at that particular time are the ones who were interviewed. There is also bias as the shoppers were interviewed outside a health food shop because this gave an impression that only those who visited the shop were more likely to be interviewed.

Parahoo (2006) argues that, only those visiting the supermarket at that time and on that day will have a chance of being selected, and accidental sampling can have implications for the data as this would not represent the views of other members of society in such an area, if the survey was to be conducted at different times of day and several times per week. Another issue to consider is that of the sample size considering that it was a Wednesday afternoon many people might have been at work and the sample size might not be representative of the population.

With regards to the age of respondents it would seem that the number of people interviewed don’t represent all the age groups accordingly. There were only 3 people from the age band 45-54 and 12 people for age band 65+ people. The author would argue that the distribution does not evenly represent all the age bands. Bowling (2002) states that, bias is possible unless the sampling method ensures that all members of the population of interest have a calculable chance of being selected in the sample.

Bowling (2002) further asserts that, a selection bias may-occur when the characteristics of the sample differ from those of wider population of interest. From the above mentioned biases at times the reliability and validity an investigation may be threatened (Bowling 2002). Bowling (2002) further argues that, although it is known that many sources of bias and error can affect social research on human beings, contamination of results is also always a threat in laboratory research in natural science.

Bowling (2002,p 177) asserts that, sampling error is the probability that any one sample is not completely representative of the population from which it was drawn. Bowling (2002, p177) further states that, the existence of a sampling error means that whatever a hypothesis is tested there is finite possibility of either rejecting a true hypothesis (type 1 error) or accepting it when it is false (type 2 error Bowling (2002, p 182) further argues, that, errors can occur for two reasons, and one is that the sampling is not carried out properly resulting in a biased sample.

Bowling (2002) states that, theoretically the errors are normally distributed with a mean of zero, so the errors balance out over all samples. The findings of this study suggests that although the respondents interviewed doesn’t evenly represent all the age bands, but a large number of the respondents 67% have used complimentary therapies, and a smaller number 32. 5% have not used complimentary therapies. It would seem that respondents from the age band 65+ were more than other age bands interviewed.

Gomm and Davies (2000) state that, samples, however, are never simply representative, they will be representative in some aspects and not in others. The findings also show that from the sex of respondents more were female (65%), and less were male with (35%) responding. The findings show that 50 percent of the respondents were of the view that complimentary therapies are most likely to work, and 5 percent thought that complimentary therapies never work. Polit and Beck (2006) assert that, a hypothesis is a prediction, usually a statement of predicted relationship between variables.

Hypothesis is a tentative prediction about the relationship between two or more variables in the population under study’ (Polit et al, 2001, p 104). They continue to suggest that it states the expected relationship between the independent variables and the dependent variables and that its use in quantitative studies tends to induce critical thinking and facilitate interpretation of the data. According to Parahoo (1997, p 126)’ to be complete and comprehensive, a hypothesis must include three components: the variables, population and the relationship between variables’.

Creswell (2003, p 94) defines the independent variable as `variables that cause, influence, or affect outcomes and states that they are also called treatment, manipulated, antecedent, or predictor variables’. He furthermore defines the dependent variables as those that depend on the independent variables, the outcomes or results of the influence of the independent variables and that its other names are criterion, outcome, and effect variables.

The two variables used are age of respondents as the independent and whether any form of complementary has been received as the dependent. From the findings it shows that people within the age band 65+ have received some form of complementary therapy more than other age bands below the age 65. Hence the hypothesis statement is’ complementary therapies are more likely to be used by those people from age 65+ class than those under this age. The null hypothesis would be that, people under the age of 65 are more likely to use complementary therapies.

How to cite this page
Choose cite format:

Statistical Package for the Social Sciences. (2017, Dec 05). Retrieved from https://primetimeessay.com/statistical-package-social-sciences/

We will write a custom essay sample onStatistical Package for the Social Sciencesspecifically for you

for only $16.38 $13.9/page
Order now