Table of contents
Academic style
View feedback on this data commentary's academic style by clicking on the underlined footnote links.
Sample BUS1100 Data analysis report
BUS1100 Assignment 1
Statistical Analysis and Report Writing
Name: John Smith
Student number: 31075698
Tutor's name: David Jones
Tutorial time: 2pm Wednesday
Table of contents
| List of Tables | i |
| List of Figures | i |
| 1 Introduction | 1 |
| 2 Data Analysis | 2 |
| 3 Conclusion | 3 |
| References | 7 |
| Appendices | 8 |
List of tables
| Table 1: Initial Summary Statistics of the Sample | 2 |
| Table 2: Measures of Central Tendency | 3 |
| Table 3: Measures of Dispersion for Manufacturing Sectors | 5 |
| Table 4: General Statistics of Manufacturing Sectors | 5 |
List of figures
| Figure 1: Histograms of Data from Different Manufacturing Sectors | 3 |
| Figure 2: Boxplot Comparison of Net Assets | 4 |
1 Introduction
This report uses a descriptive statistical method to analyse and compare the net assets ($m) of a sample of companies in four different manufacturing sectors; Sector A, Sector B, Sector C and Sector D. For the purposes of the report it is assumed that the sample of data analysed for each sector is representative of the entire population of companies. However a comparison of the data is limited by the varied number of observations for each sector. The discussion presents an analysis of the data using descriptive measures of frequency distribution, central tendency and standard deviation which reveal similar net asset figures between the four different manufacturing sectors.
2 Data analysis
To enable a comparison of the net assets (individual
data, in million dollars, $m) of the companies in the four different manufacturing
sectors a random sample from each of these populations was made (see Appendix
I)
. .............................................. [topic sentence only]
| Sector 1 ($,m) |
Sector 2 ($,m) |
Sector 3 ($,m) |
Sector 4 ($,m) |
|
|---|---|---|---|---|
| Count (no. of companies) | 73 | 45 | 19 | 50 |
| Sum | 24177.9 | 13284.6 | 5347.9 | 13941.9 |
| Range | 1570.5 | 1063.4 | 444.3 | 1746.0 |
Table 1: Initial Summary Statistics of the Sample.
(Summary statistics have been rounded off to one decimal place to maintain clarity.)
It can be seen from Table 1 that far less samples were obtained from Sector 3 than the other sectors. This could have resulted 4.1 in a significantly lower range because there was less pportunity for an individual company to have extremely high or low assets. As the number of companies in each sector increased, so did the sum of their total net assets. This is significant because it indicates that the majority of companies throughout manufacturing sectors had similar net asset values.
The general spread of net assets for each sector can be seen by allocating the raw net asset figures into frequency distributions and then histograms for each of the four manufacturing sectors 4.2. Using histograms as a graphical representation 'allows us to comprehend the shape of frequency distribution at a glance' (Harrison and Tamaschke, 1993, p. 91) and is therefore a quick and easy way to determine the asset distribution for each sector.
Figure 1: Histograms of Data from Different Manufacturing Sectors
(Note the histograms were obtained using Minitab for Windows, with 24 intervals
between the range of 0-600 $m, thus midpoints were allocated every 25$m. A number
of few extreme values have been omitted to provide a more precise representation
of the majority of the data within each manufacturing sector. For full histograms
covering the entire range of values refer to Appendix 2.)
The histograms of the manufacturing Sectors 1-4
are illustrated in Figure 1
. . . ................................................
[topic sentence only]
| Sector 1 ($,m) |
Sector 2 ($,m) |
Sector 3 ($,m) |
Sector 4 ($,m) |
|
|---|---|---|---|---|
| Median | 306.2 | 308.2 | 290.5 | 317.2 |
| Mean | 331.2 | 295.2 | 281.5 | 278.8 |
| Trimmed mean | 316.9 | 201.1 | 295.7 | 300.9 |
Table 2: Measures of Central Tendency.
As indicated in Table 2, we 4.3 can see that the median and both means for all manufacturing sectors are close to the suspected 300 $m figure. 'Measure of central tendency is defined as a single term that is considered most representative of the whole set of data' (Johnson, 1978, p. 182); thus, using a mode for this value would be inappropriate (see Appendix 1) The reason for this 4.4 is that three of the manufacturing sectors contained more than one net asset figure as the most common. In Sector 4, the minimum value was also a modal value, certainly not representing where the majority of data lies. The problem in using the mode arises from the fact that continuous data with two decimals is used for the initial net asset values, thus it is rare that two companies have equal net assets in any given sector.
Figure 2: Boxplot comparison of Net Assets ($,m)
A large difference between the median and mean for Sectors
1 and 4 respectively is apparent upon looking at Table 2 . In sector
1, the mean is far higher than the median. This is due to one company having
an extremely large net asset figure ($1530.06 $m), and many outliers above the
mean (see Figure 2). . .
..........................................[this paragraph continues]
| Sector 1 ($,m) |
Sector 2 ($,m) |
Sector 3 ($,m) |
Sector 4 ($,m) |
|
|---|---|---|---|---|
| Standard Deviation | 169.5 | 168.0 | 95.5 | 249.3 |
| Sample Variance | 28742.6 | 28239.6 | 9123.1 | 62166.3 |
| Inter-Quartile Range | 55 | 109.2 | 67.6 | 122.6 |
Table 3: Measures of Dispersion for Manufacturing Sectors
From Table 3 it can be seen that the standard
deviation for Sectors 1 and 4 are distorted by the exceptionally large and small
values within each of these sectors.
. . . ...............................................
[topic sentence only]
| Sector 1 ($,m) |
Sector 2 ($,m) |
Sector 3 ($,m) |
Sector 4 ($,m) |
|
|---|---|---|---|---|
| Minimum | -40.5 | -261.2 | -69.8 | -512.9 |
| 10th percentile | 267.2 | 188.6 | 237.6 | 157.3 |
| Lower Quartile | 289.3 | 245.4 | 260.4 | 231.4 |
| Median | 306.2 | 308.2 | 290.5 | 317.2 |
| Upper Quartile | 344.3 | 354.6 | 232.0 | 354.0 |
| 90th Percentile | 423.5 | 433.3 | 365.9 | 422.2 |
| Maximum | 1530.1 | 802.3 | 374.5 | 1233.2 |
Table 4: General Statistics of Manufacturing Sectors (calculated in Minitab)
An examination of Table 4 provides some further
interesting observations. . .
.. ............................................. [topic sentence only]
3 Conclusion
The data from four different manufacturing sectors has been analysed through the use of descriptive measures of central tendency, frequency distribution, and standard deviation. Robust statistics were also deployed to provide a better indication of net asset values within sectors with extreme values as these values can distort calculations of central tendency and dispersion. The relatively close measures of central tendency identified in the analysis suggest 4.5 that although the spread of assets between the sectors differs; sectors 1 and 3 had small spreads of assets and sectors 2 and 4 had a larger net asset range, the majority of companies across the four sectors have similar net asset values; a higher frequency of assets values at approximately 300 ($m).
References
Harrison, S. and Tamaschke, R.H.U. (1993) Statistics for business economics and management. Sydney: Prentice Hall.
Johnson, P.R. (1978) Business statistics. Sydney: MacMillan.
Luxford, K., Bedingfield, S. and Betts, J. (1998) Course notes for BUS1100 Qualitative methods for business systems. Melbourne: Monash University.
Mansfield, E. (1994 ) Statistics for business and economics and methods and applications. New York: Norton.
Appendices
Appendix 1 Random Sample of Data
Appendix 2 Histograms of the entire range of company net assets ($m) values for the four manufacturing sectors
Appendix 3 The frequency distribution of assets from the four manufacturing sectors
Tentative statements
Writing in an academic style requires the writer to make tentative, as opposed to absolute, statements. This is particularly apparent in comments which generalise from the results, provide possible explanations, or compare the results with those of other studies. In this sentence the tentative nature of the explanation has been achieved through the use of could. Other language which can be used to write tentative statements is listed below.
Probability:
can, perhaps, probably, possible, unlikely
Distancing verbs:
imply, seem, suggest, tend, appear
Frequency:
majority, overall, typical, quite, mainly, totally
Extent:
typically, generally, on the whole
See note 4.5 for another example of a tentative statement.
Linking ideas
The repetition of a key word from one sentence in a following sentence is an effective means for linking ideas between sentences. For example in this paragraph the term histogram is initially used in the first sentence and is then repeated in the second sentence. This cohesive effect can similarly be achieved by using a synonym or a pronoun for a preceding word. A smooth flow between sentences can also be created by using special link words or phrases which show specific relationships between sentences. Notice the use of therefore in this paragraph to indicate an effect. Other examples of linking words include the following.
To indicate cause:
because, since, as, for this reason
To indicate effect:
as a result, therefore, thus, consequently, thereby
To indicate contrast:
however, on the other hand, in contrast
To indicate concession:
although, however, yet, despite
To indicate an example:
for instance, for example, such as
For another example of link words in the text see 4.4.
Use of we
The use of we is frequently seen in Business Systems and other computing texts. Here the student has used we instead of alternative expressions such as "it can be seen that the median ..." or simply commencing with the phrase "the median" and deleting "we can see that ...". Generally the use of we is reserved for new or unexpected information and significant findings, it should therefore be used sparingly.
This is an example of a linking word which links ideas between sentences. The word "this" here refers to the last idea stated in the previous sentence: "using a mode for this value would be inappropriate".
This is an example of a tentative statement - "show" would be definite, "suggest" is tentative. For further information see 4.1.
Download a printable version of this page (.doc)Problems? Questions? Comments? Please provide us feedback.
