Statistical Techniques, Analysis, and Interpretation, Product Differentiation - Assignment Example

? STATISTICAL TECHNIQUES, ANALYSIS, AND INTERPRETATION, PRODUCT DIFFERENTIATION 22 March Contents ……………………………………………………………………………….3 Introduction……………………………………………………………………………...3 Statistical analysis………………………………………………………………………..5 Criticism on the data……………………………………………………………………..7 Conclusion and recommendation…………………………………………………………7 Improvement on data collection……………………………………………….………….7 Designed sheet for primary data collection………………….…………….……………..7 Why the method would improve quality…………………………………………………8 Bibliography……………………………………………………………….……………..9 Appendix ………………………………………………………….…………………….10 STATISTICAL TECHNIQUES, ANALYSIS, AND INTERPRETATION, PRODUCT DIFFERENTIATION Abstract This report discusses practical application of statistical techniques. Using secondary data, the paper investigates contents of cigarette brands with the aim of investigating product differentiation. The analysis uses both descriptive and inferential statistics to investigate product differentiation. Introduction Product differentiation is one of the emerging marketing strategies among competing business enterprises. It refers to the production of a unique commodity in the market. The underlying factor in product differentiation is that consumers must be able to distinguish the product from others in the market. The distinction may be with respect to either quality, features or even price. As a market strategy, differentiation has been used to develop brand imaging in products from organizations. Bragg explains that product differentiation is a strong marketing tool that allows firms to capture and control their customers. This is because once consumers have been attracted into a differentiated product or service they become attached to it. This in turn allows sellers to manipulate their markets for profit maximization. The underlying principle is that increasing prices of highly differentiated products does not have significant effects on the demand for such products, a feature that allows sellers to use product differentiation as a tool to market capture and profit maximization (Bragg, 2011, p. 26). Similarly, a firm can differentiate its products as a strategy for remaining competitive among other firms in the industry. Consumers’ dynamic tastes and preference may negatively shift the demand for a particular product and differentiation therefore allows organizations to develop products that will continually meet their customer’s needs. This helps firms to retain their market control and effectively manage their profit margins. A company that frequently introduces differentiated products in the market also increases its sales because consumers will always be interested in testing the new brands (Bragg, 2011, p. 26). Increasing competition in markets and the need to penetrate into the markets and maximize profits could however lead to virtual differentiation of products with the aim of taking advantage of consumers’ ignorance. Firms may for example claim that their products are unique in content mix and hence quality while such distinctions may not be real. Built loyalty and trust between organizations and their customers together with euphoria may then leads to perception and conviction over existence of differentiation that might not exist or might not be significant. A more practical approach through statistical analysis can however be employed in order to evaluate existence of differentiation as may be claimed by manufactures. Statistical tools can for instance be used to compare contents or even consumers’ utility levels through both qualitative and qualitative research initiatives. This forms ground for investigating differentiation through analysis of both descriptive and inferential statistics. Regression analysis can particularly be used to determine the relationship between properties of products such as their compositions. Similarly, analysis of variance also helps in determining significance in existence of relationship between variables (Berk, 2004, p. 21-37). This paper seeks to investigate reality in product differentiation among different cigarette brands. The paper assumes that the differentiation is based on the content. This assumption is based on the knowledge of intellectual property law that a product can only be registered if it does not infringe a patent right. Using excel, the paper will explore the relationships among properties of cigarette brands. Statistical analysis The following table summarizes the descriptive statistics as was retrieved from excel, Tar Nicotine weight Carbon monoxide Mean 12.216 0.876 0.9703 12.528 Standard deviation 5.666 0.354 0.0877 4.74 Variance 32.101 0.125 0.0077 22.4646 Range 28.8 1.9 0.3799 22 The descriptive statistics illustrate difference in the components of the different brands of cigarettes. Tar, for instance has a mean of 12.216 mg with a standard deviation of 5.666 mg and a range of 28.8 mg. This shows that the different brands of cigarettes had different composition by mass of tar. The significance of the difference is assumed from the high values of standard deviation and range relative to the mean. Similarly, nicotine content in the cigarette brands has a mean of 0.876 mg and a range of 1.9, an indication of a high level of variance in the content of the products. The same observation is made over carbon monoxide. A comparison of the components of the various cigarette brands can also be achieved through analysis of a histogram bellow. The graph depicts variation of component mix in each of the cigarrete brands, an indication of product differentiation. Similarly, regression analysis using excel offers an explorative base for analyzing the set of data. The statistical tool can be used establish existence of relationship between the weight of each cigarette and the components. An existence of a significant relationship would indicate lack of differentiation. The small value of adjusted R from the summary output, 0.143, shows that the regression model explains less than fifteen percent of the data used in the analysis. The model would therefore not be suitable for explaining the relationship between weights of cigarettes and the corresponding masses of their components. The high significance value as reported in the ANOVA table, relative to the significance level (0.05) also illustrates lack of significance relationship between the variables. A consideration of the coefficients of variables confirms this observation due to the high probability values relative to the predetermined level of significance, 0.05. This confirms ‘non-existence’ of a relationship between weights of the cigarettes and the composition by mass of the components. Criticism on the data A possible criticism of the data is its reliability. Having been obtained from a secondary source, the audience may be keen biasness in the collected information. The data was however retrieved from an academic institution, Illinois University. This supports its reliability based on the credibility of its source. Conclusion and recommendation Lack of significant relationship indicates that the different brands of cigarettes are made from uncorrelated formulas. The products are therefore significantly differentiated. Further research should however be carried out to investigate the status of differentiation among cigarette brands and even other products. This will facilitate control on counterfeit goods. Improvement on data collection An improvement to the data collection would be achieved through personal observation that would ensure primary source of data. Samples of each cigarette brand would be obtained from each outlet and compared for average values. Determination of an appropriate sampling method would be necessary (Ott and Longnecker, 2008, p. 17- 30). Designed sheet for primary data collection The following sheet would be used to record primary data for mean weights from different outlets. brand Shop outlet Weight (g) Tar (mg) Nicotine (mg) Carbon Monoxide(mg)           Why the method would improve quality The method would improve quality because it will employ triangulation of data to collect information from different sources for comparison. This will improve reliability and precision in research. Bibliography Berk, R. (2004). Regression analysis: a constructive critique. London, UK: SAGE Bragg, S. (2011). The New CEO Corporate Leadership Manual. New York, NY: John Wiley & Sons Ott, L. and Longnecker, M. (2008). An introduction to statistical methods and data analysis. California, CA: Cengage Learning Appendix 1: Secondary data used Brand Tar (mg) Nicotine (mg) Weight (g) Carbon Monoxide(mg) Alpine 14.1 0.86 0.9853 13.6 Benson & Hedges 16 1.06 1.0938 16.6 Bull Durham 29.8 2.03 1.165 23.5 Camel Lights 8 0.67 0.928 10.2 Carlton 4.1 0.4 0.9462 5.4 Chesterfield 15 1.04 0.8885 15 Golden Lights 8.8 0.76 1.0267 9 Kent 12.4 0.95 0.9225 12.3 Kool 16.6 1.12 0.9372 16.3 L & M 14.9 1.02 0.8858 15.4 Lark Lights 13.7 1.01 0.9643 13 Marlboro 15.1 0.9 0.9316 14.4 Merit 7.8 0.57 0.9705 10 Multi Filter 11.4 0.78 1.124 10.2 Newport Lights 9 0.74 0.8517 9.5 Now 1 0.13 0.7851 1.5 Old Gold 17 1.26 0.9186 18.5 Pall Mall Light 12.8 1.08 1.0395 12.6 Raleigh 15.8 0.96 0.9573 17.5 Salem Ultra 4.5 0.42 0.9106 4.9 Tareyton 14.5 1.01 1.007 15.9 TRUE 7.3 0.61 0.9806 8.5 Viceroy Rich Light 8.6 0.69 0.9693 10.6 Virginia Slims 15.2 1.02 0.9496 13.9 Winston Lights 12 0.82 1.1184 14.9 Appendix 2: summary output table Regression Statistics Multiple R 0.500336 R Square 0.250336 Adjusted R Square 0.143241 Standard Error 0.081196 Observations 25 Appendix 3: ANOVA table ANOVA   df SS MS F Significance F Regression 3 0.046232 0.015411 2.337514 0.102789 Residual 21 0.138449 0.006593 Total 24 0.184681       Appendix 4: table of coefficients   Coefficients Standard Error t Stat P-value Intercept 0.864079 0.061532 14.04273 3.81E-12 X Variable 1 0.001161 0.018006 0.064459 0.949214 X Variable 2 0.110888 0.220102 0.503804 0.619643 X Variable 3 -0.00041 0.012255 -0.03358 0.973527 Read More