Review Of Statistical Disclosure Control Techniques: Data Presentation Methods

Evidence from literature

Discuss about the Review Of Statistical Disclosure Control Techniques.

In today’s world collection of data and its application has become a major part of any organization. Depending on methods of collection data are classified in two groups- primary data (collected from direct source) and secondary data (collected from indirect source). Reliable data source is one primary aspect of data collection. An equally important thing is presentation of data. The paper considers different data presentation techniques and their application in real world.  The three types of data presentation methods discussed are textual, tabulation and graphical. All the three forms have their respective merits and demerits. Despite some demerits, appropriate data presentation technique is chosen after scrutinizing the statistical data.

Data and information play an important role in all part of academic studies, commercial, business and marketing activities. After collection and classification of data, the next important step is to present the collected data in an appropriate form. The raw data collected should be processed and presented in suitable format to make further use of the data. After collection of data, it should be classified in different groups to make data presentation easy. The four major types of classification include classification on qualitative basis, on quantitative basis, on time basis and on geographical basis (Weissgerber et al. 2015). Qualitative classification refers to classification of total population according to attributes like religion, sex and occupation. Classification of population based on age, industry classification according to number of person employed are examples of quantitative classification. When data are arranged according to time then it is known as time series data. Classification can also be done using geographic characteristics. Quantitative data can further be classified into frequency and non-frequency group. These different forms of data need proper means of presentation (Chambers 2018). It can be presented in textual format or using tables or using graphs. Which particular form is to be used that depend on nature and size of data. For a very small data, a textual format can be the most appropriate form of data while for large data tabulation or graphs are the most suitable means of data presentation (Anderson et al. 2016). The next two sections discuss standard methods of data presentation with practical application of these techniques.

Presentation of data is broadly classified into three major categories

• Textual Presentation
• Tabulation
• Diagrammatic presentation

It is a very common method of presenting statistical data. The textual presentation refers to presenting data using paragraphs of text. This technique is employed in preparing official report where plans, activities or programs are described in words. In such reports relevant numerical figures are inserted in between the texts (Ross 2017). This is the simplest technique of data presentation. There is no hard and fast rule for presenting data in textual format. One thing that should be taken care of is the logical sequence of the data and clarity of the presentation. The data should be presented using precise and brief text.

Tabulation is a compact form of data presentation. This involves a systematic presentation of data in the form of a table structure comprising relevant rows and columns. There are two common form of tabulation

Different methods of data presentation

i)Simple Tabulation: In a simple table only one characteristic of data can be presented. In the simple tabulation other characteristics of data are left out.

ii)Complex Tabulation: With complex table one can present different dimension of data in a composite form.

There may be single, double, treble or manifold tabulation. The single tabulation answers one or more independent questions (Najafabadi et al. 2015). Whereas in double, treble or manifold tabulation contains two, three or more subdivisions according to characteristics of presented data and can address independent as well as mutually dependent questions.

Different forms of charts, maps, pictures, are effective and attractive means of statistical data presentation. The use of diagrams in presenting data has the advantage of readily capturing some feature of the exhibited data. The primary objective of using diagrams is to emphasize relative position of different subdivision (Copland and Creese 2015). There are different types of diagrams that can be used for present data depending on the nature of statistical data. The commonly used diagrams are the followings

• Line diagram or graph
• Bar Diagram
• Pie diagram
• Pictogram
• Histogram

The easiest and common form of diagram is line diagram. Line diagram is particularly applied in field of commerce and business where data are presented with respect to time. The line diagram depicts relationship between two variables. A straight line implies linear relationship while quadratic or polynomial relationship is shown by means of a curve. For construction of line diagram, two co-ordinates are taken (Bendig,et al. 2017). The horizontal axis measures time while the vertical axis represents the corresponding variables. After selecting the co-ordinates suitable measures of scale are chosen to plot the data points. The data of the chosen variable is then plotted against time to obtain co-ordinates of points. The points thus obtain are then joined and the smooth curve obtained is called line diagram.

In this form of diagrammatic presentation there are number of equidistant rectangles termed as bars, each corresponds to some specific category of available data. The bars have a common width and are drawn on the base line. The base line represents various categories. The height or length of bars shows value corresponding to each category. Bar diagrams are of two types- vertical bar diagram and horizontal bar diagram. The vertical bar diagram is applicable to time series data or for the data classified according to values of the variable. Horizontal bars are used to present data that contains attributes. For each of these bar diagram there are again grouped bar diagram, paired bar diagram and subdivided bar diagram. In order to show comparison from two or more statistical group of data grouped bar diagram is used (Matthews, Harel and Aseltine Jr 2017). The subdivided or component bar diagram shows comparison among different components of the bars and show relation between different part and as a whole. In the paired bar diagram there are several pairs of horizontal bars extending in opposite direction.

Pie diagram is an appropriate form of diagrammatic representation for presenting data exhibiting relative sizes of different parts as a whole. In pie diagram, a circle is partitioned to present different parts of the data. The relative values of each part is first converted to percentage form of the whole and then converted to respective angles. Each sector of the pie diagram represents concerned percentage of the part as a whole (Nadeem, Zafar and Zahid 2015). The area enclosed by the circle is regarded as 100. As circle represents angles measures 360⁰, each part is multiplied by 3.6 to find out corresponding angles. Pie diagram is very useful when data are subdivided in a number of categories and the researcher is interested in comparing various categories or between different parts or as a whole.

Textual Presentation of data

Pictogram comprises rows of picture symbols, all having equal size. Each symbol is representative of a definite numerical value. In case there is a fraction of value, then appropriate proportion of the symbol is shown. Pictogram is used to present time series data (Peck, Olsen and Devore 2015). For each time period, there is a row of pictures. Pictograms are also very useful in presenting statistical data containing attributes.

Histograms are used to present data containing grouped frequency distribution. It comprises of several adjoining rectangles drawn on the base line having areas proportional to respective class frequencies (Jin,et al., 2016).

In this section a comparative analysis is made among different methods of data presentation. Each method has some advantages and disadvantages over others. The selection of appropriate method depends on the nature and size of the data.

Starting with textual presentation, it is the easiest way of presenting data. This mode of data presentation has an appeal to people with a literacy bent of mind. The researchers can draw attention of people towards a certain point which appears to be important (Sharma et al. 2017). This is explained with following example of textual representation.

women and 1576 women participated in an opinion poll for a certain government measures. 1560 persons of whom 1176 were male, voted against the measure. Overall, 2025 persons were voted for the measure, while 365 women remain indifferent…..”. This is how textual representation in combination of text and figures presents information regarding any events.

Despite being the simplest form of data presentation technique, it becomes a time consuming and cumbersome method for large data. One has to read the text for understanding the data. This method is not suitable for presenting a large mass of data. It is difficult to draw comparison among data point from textual presentation. Visually the textual presentation seems boring and monotonous. One primary purpose of data collection is to analyze the data. However, with textual presentation the data are not ready for statistical analysis.

A comparatively better form of data presentation technique is tabulation. A good table should comprise of following parts

Title: The title contains brief description of the table and is shown at the top of the table.

Stub: The extreme left part of the table is called Stub. This part contains description of rows.

Caption or box head: This is the upper part of the table showing description of columns and sub columns. The whole upper part including caption, measurement units and column number is known as box head.

Body: This contains the data points or figures.

Unlike textual presentation, with tabulation large data can be presented in a composite form. The tabular form of presentation any errors or omissions of the data can be easily identified. Tabulation presents large numerical data easily. A table containing each of the above discussed part is easy to comprehend and data from the table can be used for statistical analysis. Complex data can also be presented with subdivision of the tables (Mardia 2014). Tables clearly specify characteristics of data and also enable reader to compare information. Data can be understood with a much less time as compared to textual presentation.

Tabulation

Tabulation though is a superior technique than textual but is has disadvantages too. For a layman it is not possible to comprehend data from tabulation. It fails to create a lasting impression to the reader.

The diagrammatic representation is the most appealing form of data representation technique. There are different types of diagram applicable to different types of data.

Unlike tabulation, when data are presented by the means of graph then even a layman can understand the data. It requires very less time convey information to people in general. One can have an idea about significance of the presented data at a glance (Mertler and Reinhart 2016). Graphs crate a long lasting impression for the readers. Two or more series of data can be compared with the use of graphs.

However, diagram fails to represents any detailed information that tabulation or textual representation. As diagrams represent approximated value, data precisions are often lost. For construction of diagram one need to devote sufficient time. Moreover, a diagram present only limited information.

When the size of data is not too large and there is no difficulty in understanding data by go through simple texts, then textual presentation is an appropriate form to be used. In real world, in short presentation often minor data are presented in simple text format. For example, nearly 60% people in America are employed in different industries. This minute information nether require a table nor a graph. For large data and data indicating several characteristics tabulation is used (Kosara 2016).  In academics, tables are used to present statistics of students’ score. For example, suppose data on result of annual examination of BA and B.Sc. program is in a particular year is given. Then tabular form is used to represent this information. The table looks as follows

Diagrams are the most commonly used means of data presentation. Line char is mostly used for showing trend of a variable over time. For example, growth of population, sales of firms’ overtime, import, export and other important variables are presented using line charts. For showing comparison among countries, across genders and any other form of comparison bar graphs is used (Gupta 2016). Similarly, pie chart and histograms are also used wherever found necessary.

Conclusion

The three data presentation techniques discussed are textual presentation, tabulation and diagrammatic presentation. With textual presentation data are presented in paragraphs with important figures inserted in between words. This form in very limited in use as it cannot represent large data. In tabulation data having several variables or attributed can be presented compositely. Tabulation has wider application than textual but still its use is limited to a certain extent. The widely used data presentation technique is the graphical approach. Different graphs (line, bar, pie, pictogram, histogram) are used for different purposes and provide a wider understanding of the data by only visualizing the graph.

Reference list

Anderson, D. R., Sweeney, D. J., Williams, T. A., Camm, J. D. and Cochran, J. J., 2016. Statistics for business & economics. Nelson Education.

Bendig, K., Eby, T., Dowling, C.G., Clouse, R., Sheth, K., Suh, E.Y. and Huang, D., Microsoft Technology Licensing LLC, 2017. Smart card presentation of tabular data from collaboration database. U.S. Patent Application 15/198,375.

Chambers, J. M., 2018. Graphical methods for data analysis. CRC Press.

Copland, F., and Creese, A., (2015). Linguistic ethnography: Collecting, analysing and presenting data. Sage.

Data.worldbank.org. (2018). Australia | Data. [online] Available at: https://data.worldbank.org/country/australia [Accessed 22 Jan. 2018].

Gupta, S.C., 2016. Fundamentals of Statistics. Himalaya Publishing House.

Jin, Y., Liao, J., Shao, Y., Yin, G. and Zheng, J., International Business Machines Corp, 2016. Interactive threshold setting for pie charts. U.S. Patent 9,424,670.

Kosara, R., 2016. Presentation-oriented visualization techniques. IEEE computer graphics and applications, 36(1), pp.80-85.

Mardia, K.V., 2014. Statistics of directional data. Academic press.

Matthews, G.J., Harel, O. and Aseltine Jr, R.H., 2017. A review of statistical disclosure control techniques employed by web-based data query systems. Journal of Public Health Management and Practice, 23(4), pp.e1-e4.

Mertler, C.A. and Reinhart, R.V., 2016. Advanced and multivariate statistical methods: Practical application and interpretation. Taylor & Francis.

Nadeem, M.F., Zafar, S. and Zahid, Z., 2015. On certain topological indices of the line graph of subdivision graphs. Applied Mathematics and Computation, 271, pp.790-794.

Najafabadi, M.M., Villanustre, F., Khoshgoftaar, T.M., Seliya, N., Wald, R. and Muharemagic, E., 2015. Deep learning applications and challenges in big data analytics. Journal of Big Data, 2(1), p.1.

Peck, R., Olsen, C. and Devore, J.L., 2015. Introduction to statistics and data analysis. Cengage Learning.

Ross, S.M., 2017. Introductory statistics. Academic Press.

Sharma, M.C., Saxena, D. and Singh, V., 2017. Unit-8 Statistics: Averages, Graphic Representation and Classification of Data.

Weissgerber, T. L., Milic, N. M., Winham, S. J. and Garovic, V. D., 2015. Beyond bar and line graphs: time for a new data presentation paradigm. PLoS biology, 13(4), e1002128.