QuantitativeAnalysis: Descriptive Analysis
Incarrying out statistical analysis, there is need to understand someof the aspects of the data that would help in proper analysis of thesame. datasets are usually grouped into categories in accordance withtheir nature. the two major groupings of datasets are categoricaldataset and the continuous dataset. In the categorical dataset, thedata is usually separated into classes which represents a specifictype of information. They are deemed to be qualitative or discretevariables. The types of variables that falls in the categorical groupare such as the nominal, dichotomous and the ordinal variables(Powers,Xie, & Xie, 2011).The nominal variable contains the categories of datasets thataccentuates more than two categories of data types. For example, wecan have a country’s population organized into western, eastern,central, northern and eastern.
Dichotomousdatasets on the other hand contains two distinct categories of datawhich are to be analyzed for example, the categorization of populaceinto male and female. The third categorical datasets represented hereis the ordinal variables. In this context, the variables are justlike the nominal data, its only that these variables are ranked inorder of importance. For example, on the collection of the opinion ofpersons of the elections whether a presidential aspirant would win,one may say yes, mostly likely, not likely or no. These sets of dataare collected and summarized to form the ordinal data variables(Powers,Xie, & Xie, 2011).Continuous variables on the other hand are described as quantitativevariables. they can be further grouped into interval and ratiovariables. Interval variables are those that can be measured along acritical continuum while maintaining their representation in anumerical value. For example, the measurement of temperature indegree Celsius. Ratio variables are essentially interval variablesits only that zero (0) means there is no measurement (Rosenberg,Adams, & Gurevitch, 2014).This paper looks into the differentiation between the categorical andcontinuous dataset in the practical sense through descriptivestatistical analysis.
Inorder to understand the context both the categorical and thecontinuous datasets, a descriptive statistics on a section of theAfrobarometer dataset 4(http://www.icpsr.umich.edu/icpsrweb/ICPSR/studies/33701)was done and analyzed in this paper to accentuate the clear cutdifference. In this prospect, the descriptive statistics of the“within country weight’ and that labelled, ‘urban or ruralsampling unit’. The dataset for the ‘urban or rural sampling unitis a nominal variable which indeed is a categorical variable. The‘within country weight’ is a continuous ratio variable whichconnotes an investigation within a continuum of weight (Rosenberg,Adams, & Gurevitch, 2014).From the context of this information the descriptive statisticsresults for both the categorical and the continuous datasets are asshown in the SPSS output below.
GET
FILE=`C:UsersuserDownloadsmerged_r4_data.sav`.
DATASETNAME DataSet1 WINDOW=FRONT.
DESCRIPTIVESVARIABLES=URBRUR
/SAVE
/STATISTICS=MEANSTDDEV VARIANCE RANGE MIN MAX SEMEAN KURTOSIS SKEWNESS.
DescriptiveStatistics Categorical – Nominal
[DataSet1]C:UsersuserDownloadsmerged_r4_data.sav
Descriptive Statistics 

N 
Range 
Minimum 
Maximum 
Mean 
Std. Deviation 
Variance 
Skewness 
Kurtosis 

Statistic 
Statistic 
Statistic 
Statistic 
Statistic 
Std. Error 
Statistic 
Statistic 
Statistic 
Std. Error 
Statistic 
Std. Error 

Urban or Rural Primary Sampling Unit 
27713 
1 
1 
2 
1.62 
.003 
.485 
.236 
.496 
.015 
1.754 
.029 

Valid N (listwise) 
27713 
DESCRIPTIVESVARIABLES=Withinwt
/SAVE
/STATISTICS=MEANSTDDEV VARIANCE RANGE MIN MAX SEMEAN KURTOSIS SKEWNESS.
DescriptiveStatistics – Continuous scale
[DataSet1]C:UsersuserDownloadsmerged_r4_data.sav
Descriptive Statistics 

N 
Range 
Minimum 
Maximum 
Mean 
Std. Deviation 
Variance 
Skewness 
Kurtosis 

Statistic 
Statistic 
Statistic 
Statistic 
Statistic 
Std. Error 
Statistic 
Statistic 
Statistic 
Std. Error 
Statistic 
Std. Error 

Within country weight 
27713 
7.5054 
.0058 
7.5113 
.999987 
.0032494 
.5409397 
.293 
2.042 
.015 
10.986 
.029 

Valid N (listwise) 
27713 
Fromthe above SPSS output, it comes out clearly that there is indeed abig difference between the categorical and the continuous output. Inthe first instance, the continuous descriptive statistics output isrepresented in numerous decimals while that of the categoricalvariable are shown without decimals (Rosenberg,Adams, & Gurevitch, 2014).The range for rural urban sampling unit (categorical) for example is1 while that of the within the country weight (continuous) is 7.5054.The notion of being continuous therefore makes this data to berepresented with numerous decimals to connote the continuity(Campbell,Julious, & Altman, 2015).According to the descriptive statistics as well, the categoricalvariable (urban sampling unit) is seen to have a minimal standarderror of 0.003 while that of the continuous variable (within countryweight) is a little higher at 0.0032494. These representations bringon board the clear difference between these sets of data variables(Powers,Xie, & Xie, 2011).The categorical data on the sampling units connotes the categories ofgroups that were used as sampling units. For example, the 1represented the urban while 2 represented the rural. On the accountof continuous variables, the within the country weight the continuumhere is the ranges of weights which falls within that of thestipulated weights of the country (Campbell,Julious, & Altman, 2015).
Conclusively,In the categorical dataset, the data is usually separated intoclasses which represents a specific type of information. They aredeemed to be qualitative or discrete variables. The types ofvariables that falls in the categorical group are such as thenominal, dichotomous and the ordinal variables. Continuous variableson the other hand are described as quantitative variables. they canbe further grouped into interval and ratio variables (Rosenberg,Adams, & Gurevitch, 2014).Interval variables are those that can be measured along a criticalcontinuum while maintaining their representation in a numericalvalue. For example, the measurement of temperature in degree Celsius.The main representation here in the descriptive statistics here isbased on the number of decimals bestowed upon each statistic. Forcategorical variables, the datasets are represented with minimal orno decimals while with continuous variables, numerous decimals arequite essential.
References
Powers,D. A., Xie, Y., & Xie, Y. (2011). Statisticalmethods for categorical data analysis(Vol. 106). New York: Academic Press.
Rosenberg,M. S., Adams, D. C., & Gurevitch, J. (2014). MetaWin:statistical software for metaanalysis.Sunderland: Sinauer Associates.
Campbell,M. J., Julious, S. A., & Altman, D. G. (2015). Estimating samplesizes for binary, ordered categorical, and continuous outcomes in twogroup comparisons. BMJ:British Medical Journal,311(7013),1145.