Abstract
Business strategy is the strategy which the company uses to compete with their rivals in the same industry. Business strategy is also considered as the spine of the business because it dominates other operations in the company and helps create competitive advantages over their competitors. Due to these very important roles of the business strategy, we decided conduct this research paper with the aim to test whether choosing different types of business strategies will lead to difference in the performance of the companies or not. This research paper is based on the data from quarterly financial reports of the three companies in the computer industry in US from 1999 to 2011. With the findings of the research, the paper draws a conclusion that the impacts of different strategies are not the same, so that the managers should choose the most appropriate and profit-maximizing strategy for their company. Moreover, this result of the test will give incentives to the firms to put forward further test to find out the most effective strategy, which then can be recommended to Vietnam’s businesses.

I. Scenario
Computer industry is a very young industry in the world and it was not until the late 1990s that it really developed rapidly. In spite of this fact, the computer industry has gained a lot of significant achievements, opening a new era for the human-beings. It is also considered as an industry that has the fastest growth rate with a lot of new innovative products being produced every minute. As well as this, the number of companies operating in this field also increases considerably. Specifically, there are more than 2000 computer companies worldwide, creating a very competitive and hostile environment to work in. On the contrary to the rapid development in the world, the computer industry in Vietnam is still so weak that it can not keep up with the pace of changing technology in the world. Thus, most of the computers used in Vietnam are still imported from abroad. Therefore, in order to survive and prosper in such environment, Vietnamese companies should have to choose the most appropriate business strategy to match with the organization’s strengths and core competencies, so that they can compete with their foreign rivals. However, Vietnamese companies often confuse when choosing the strategy because there are many types of strategy. Realizing this fact, we think that it is necessary to do a research to find out whether different strategies will have different influence on the performance of the companies or not and then recommend the most effective business strategy to Vietnamese companies. In fact, this research paper only refers to three major types of business strategy: cost leadership, differentiation and customer service. When an organization competes on the basis of having the lowest cost of production in its industry, it should follow a cost leadership strategy. But if company offers unique products that are widely valued by customers, it uses differentiation strategy. Company can choose customer service strategy to focus on satisfying the demand of customer well. Our research is based on the figures gathered from the quarterly financial reports of 3 companies in US: IBM, Dell, Sony, aims to thoroughly test whether differences exist among the three strategies in terms of their effectiveness by applying single-factor analysis of variance with three independent samples (one-way ANOVA). Concurrently, this piece of research also makes logical analysis about collected data and show the most proper strategy for firm to run the business.

II. Research methodology
1. Population and sample: • Populations: we randomly choose three computer corporations: IBM, Sony and Dell among 9 computer corporations in US. The populations are 48 quarterly sales from 1999 in each of these three computer corporations (3 treatments). • Samples: ＋Sampling method: We choose all quarterly sales from 1999 (48 quarterly sales) in each of these three computer corporations. We do not mention the figures before 1999 because the computer industry has only developed rapidly since this year.

＋Sample size: Among all quarterly sales we choose, we randomly take samples size of 15 quarterly sales for each treatments.

• Factor : The factor that defines the populations is the business strategy that has 3 levels ( cost leadership, differentiation and customer service) • Responses in our test are actual value of sales of the 3 corporations.

2. Research method:
Corporations in different samples have used different business strategies (factors). To be more specific, Sony, IBM and Dell have used cost leadership, differentiation and customer service respectively. All of them use only one strategy to operate their business. Through their annual financial reports on their websites, we collect the data of quarterly sales each year from 1999 and take samples. We recorded all data by excel and used them for research purpose. The raw material would appear in Appendix.
We decide to choose the one-way ANOVA for the experiment because the purpose of this test is to examine whether the turnover is influenced by the business strategy or not. In addition, the populations are also satisfied the assumptions of ANOVA test: data is quantitative and the samples we collected are large enough to become appropriately normal. Moreover, we can realize that all of these corporations have quite similarity in their size and their ages. As an evidence for this, all three companies has around 500 branches in about 120 countries, more than 300,000 employees and all of them has operated business in this area for approximately 30 years. Thus, the size and the age of company do not affect the differences in sales of 3 companies much, therefore they do not affect results remarkably.
For the reason that there is an inverse relationship between the probability of making a Type I error (alpha) and the probability of making a Type II error (beta), for a given sample size, any decrease in alpha results in an increase in beta. Therefore, we decide to choose the reasonable value of alpha equals to 0.05.

3. Conclusion
Using ANOVA test we can conclude whether 3 population means are different or not given chosen α.There are two possible results: • If we reject the null hypothesis or there is enough evidence to conclude that there are differences among the 3 population means, we will continue to conduct a test to find out the most effective factor to use. • If we do not reject the null hypothesis. It means that there is not efficient evidence to conclude that there is any factor of the three shows more effectively than the others. In this case, we choose the smallest factor regarding to economical reason.

III. Analysis and Discussion: 1. Analysis: * The problem question: “Can we conclude at 5% significant level that a difference exists between the sales generated from using different business strategies.” - The problem objective is to compare 3 populations (quarterly sales of 3 companies: IBM, Dell, Sony, which use different strategies: Differentiation, Customer service, Cost leadership respectively). - The data is quantitative. - The parameter of interest is the central location, the mean quarterly sales of 3 companies. Therefore, the proper technique to be used is the Analysis of Variance. *Checking assumptions: - Firstly, the F-test of the analysis of variance requires that the random variable is normally distributed. This requirement is easily checked by drafting the histograms for each sample. From the histograms below, we can see that the three populations are approximately normal, so the requirement is satisfied.

[pic] [pic] [pic] - Secondly, the population variances must be equal: [pic]
From the data shown in appendix, the sample variances are calculated as follows: [pic] = 8.2

[pic] = 8.7

[pic] = 8.65

Obviously the three sample variances are approximately equal to each other. This allows us to assume that the population variances are equal. • Samples are independent.
Hypotheses:
[pic]
[pic] At least two means differ
Test statistic:
[pic]
Significance level: α = 0.05
Decision rule: Reject H0 if F > Fα,k-1,n-k = F0.05,2,42 = 3.23 [pic]
Value of the test statistic:
Using the application of Microsoft Excel, the ANOVA table is conducted and interpreted below:

|SUMMARY | | | | |
|Groups |Count |Sum |Average |Variance |
|24.12 |14 |312.66 |22.33286 |8.597684 |
|6.14 |14 |193.37 |13.81214 |5.140464 |
|16.58 |14 |217.32 |15.52286 |9.230468 |

|ANOVA | | | | | | |
|Source of Variation |SS |df |MS |F |P-value |F crit |
|Between Groups |568.891 |2 |284.4455 |37.15228 |9.29E-10 |3.238096 |
|Within Groups |298.592 |39 |7.656205 | | | |
| | | | | | | |
|Total |867.483 |41 | | | | |

As the table shows, the value of the test statistic is:
F = 5.33 > F0.05,2,42 = 3.23 ( Reject H0.

Conclusion: There is enough statistical evidence to infer that differences exist among the average quarterly sales of three companies which use different business strategies at 5% level of significance.

2.Discussion: - As being mentioned above, a question arose before the test is that whether different types of business strategies will have different effects on the performance of the companies (reflected through their sales). Basing on the analysis of the findings, we can conclude that three types of strategies (cost leadership, differentiation, focus on customer service) do not have the same impacts on the sales of three famous computer-manufacturing company (Sony, IBM, Dell) at 5% level of significance. It is a helpful information for manager because it reflects that business strategy is one of the important factors affecting their sales, so if the company are experiencing a decline in sale, manager should consider their strategies again to make sure that whether the strategy is appropriate or not. Then managers can take some actions to make necessary adjustment to their strategies to improve sales. For example, they can make a combination of two most effective strategies to increase their sales, such as utilizing both the cost leadership and the customer service rather than using just only current strategy. - Moreover, this difference in the effectiveness of different strategies may give researchers incentives to conduct further tests to find out which is the best strategy. Next, we can apply the practices of the most effective strategies for the computer industry in Vietnam, which is still very young and weak. For instant, if further research prove that the strategy which focusing on customer services is the most effective one, generates the highest profit, Vietnamese computer companies can consider this strategy as one direction to improve their company’s performance. To be more specific, Vietnamese company can imitate the practices of Dell – a very successful corporation in using this strategy. Like Dell, they can do all customer service online, from which customer can choose any parts that they prefer for their new computer through internet, then their orders are transmitted directly to the manufacturing department, so they can produce a computer that can satisfy the customer’s need quickly. - Additionally, the features of the products are affected significantly by the business strategy, such as cost leadership strategy will lead to low-price product or differentiation strategy will help create innovative products that may be more expensive but have some outstanding features. Therefore, it can not be denied that the difference in sales generated from different strategies will reflect the trend of the customers’ tastes. For instance, if the sales of products of company that follow differentiation strategy is highest, it means that people nowadays seems to be more favorable in the quality of computers, they may prefer computer with more the innovative applications, so they are willing to pay more for the computer. Knowing this trend, computer-manufacturing companies can have more precise plan of production for future by focusing on features that customers value most in order to attract customers and improve their profit.

IV. Limitation.
There are some limitations of our tests that might distort the conclusion of the test to some certain extent. - Firstly, it is obvious that there are some other factors that can affect the company’s sale rather than just business strategy and the size of the business. For example, the sales may depend on the general economic situation, the effectiveness of the management board or the quality of the staffs, etc. However, the business strategy still has the greatest impact because business strategy can influence all the operation, decisions of managers and staffs or even the future plan of the company. Hence, we still can conclude that the performance of the company (reflected in the sales) is influenced mostly by its business strategies despite some small distortion caused by other factors. - Secondly, Ho is rejected; however, we can not be absolutely sure that Ha is true. Thus, we can not 100% confident that the business strategies affect differently on the sales. We just prove there is evidence so strong against Ho and in favor of Ha. If we reject Ho when they have proved that Ho is true, this mean we have made Type I error and this probability is equal to 5%.

V. Conclusion and recommendation for further research:

With the aim of considering whether or not the business strategy affects the revenues of companies, we conducted this research and some conclusions can be drawn as listed below:
Firstly, the ANOVA test helps us determine whether differences exist among population means in an effective way. Therefore, we can use it popularly in our daily life, specially in business field. In this situation, we choose single-factor ( one way) analysis as the most appropriate technique to carry out the test rather than the multifactor model. It is really important for us to apply the suitable approach to reach the accurate results.
Next, we have strictly followed all steps from checking assumptions to drawing final conclusion to practice ANOVA. Actually we can solve by hand or employing different software test but in our research, we applied Microsoft Excel to analyze data in order to perform the test faster and more accurate and still give priority to the process of doing the test.
Finally, after sufficient process of doing this project as described above, we can strongly conclude that the data provide enough evidence at 5% significance level to infer that the business strategy has strong impacts on the revenues of the companies.

VI. References:

http://www.ibm.com/us/en/sandbox/ver1/ http://www.dell.com/ http://www.sony.com/ http://www.wikinvest.com/stock/Dell_(DELL)/Data/Revenue http://www.wikinvest.com/stock/International Business Machines_(IBM)/Data/Revenue http://www.wikinvest.com/stock/Sony_(SNE)/Data/Revenue VII. Appendix:

Appendix 1: List of computer companies using 3 types of business strategies in US

|Cost |Differentiation |Customer |
|leadership | |service |
|Acer |IBM |HP |
|MIS |Apple |Dell |
|CMI | |Asus |
|Compuware | | |
|Sony | | |

Appendix 2: Population list of IBM (quarterly sales from 1999 to 2010)

|IBM | | | | | |
| |Annual |Q1 |Q2 |Q3 |Q4 |
| |1999 | | | | |
| | |$20.32 |$21.91 |$21.14 |$24.18 |
| |2000 | | | | |
| | |$19.35 |$21.65 |$21.78 |$25.65 |
| |2001 | | | | |
| | |$21.04 |$21.57 |$20.43 |$22.83 |
| |2002 | | | | |
| | |$18.55 |$19.65 |$19.82 |$23.68 |
| |2003 | | | | |
| | |$20.07 |$21.63 |$21.52 |$25.91 |
| |2004 | | | | |
| | |$22.25 |$23.15 |$23.43 |$27.46 |
| |2005 | | | | |
| | |$22.91 |$22.27 |$21.53 |$24.43 |
| |2006 | | | | |
| | |$20.66 |$21.89 |$22.62 |$26.26 |
| |2007 | | | | |
| | |$22.03 |$23.77 |$24.12 |$28.87 |
| |2008 | | | | |
| | |$24.50 |$26.82 |$25.30 |$27.01 |
| |2009 | | | | |
| | |$21.71 |$23.25 |$23.57 |$27.23 |
| |2010 | | | | |
| | |$22.86 |$23.72 |$24.27 |$29.02 |

Appendix 3: Population list of Dell (quarterly sales from 1999 to 2010)

|Dell | | | | | |
| |Year |Q1 |Q2 |Q3 |Q4 |
| | | | | | |
| |2010 |14.90 |14.87 |15.53 |15.39 |
| |2009 |13.43 |12.34 |12.76 |12.90 |
| |2008 |15.99 |16.08 |16.43 |15.16 |
| |2007 |14.47 |14.72 |14.78 |15.65 |
| |2006 |15.18 |14.22 |14.21 |14.42 |
| |2005 |13.46 |13.39 |13.43 |13.91 |
| |2004 |11.51 |11.54 |11.71 |12.50 |
| |2003 |9.74 |9.53 |9.78 |9.62 |
| |2002 |8.06 |8.07 |8.46 |9.14 |
| |2001 |8.67 |8.03 |7.61 |7.47 |
| |2000 |6.80 |7.28 |7.67 |8.26 |
| |1999 |5.17 |5.54 |6.14 |6.78 |

Appendix 4: Population list of Sony (quarterly sales from 1999 to 2010)

|Sony | | | | | |
| |Year |Q1 |Q2 |Q3 |Q4 |
| | | | | | |
| |2010 |18.14 |18.77 |20.75 |27.20 |
| |2009 |9.80 |16.58 |18.55 |24.04 |
| |2008 |28.03 |18.67 |19.48 |23.77 |
| |2007 |18.26 |16.07 |18.11 |25.08 |
| |2006 |15.79 |15.17 |15.71 |21.91 |
| |2005 |13.88 |14.18 |15.07 |20.06 |
| |2004 |18.58 |14.93 |15.34 |20.86 |
| |2003 |14.77 |13.37 |16.19 |21.71 |
| |2002 |13.84 |14.47 |14.67 |19.23 |
| |2001 |11.85 |13.21 |15.02 |17.33 |
| |2000 |13.76 |14.87 |15.70 |18.35 |
| |1999 |11.28 |12.25 |15.26 |18.78 |

Appendix 5: Sample list of the three treatments:

|IBM |Dell |Sony |
|Differentiation |customer service |cost leadership |
|$24.12 |6.14 |16.58 |
|$19.82 |16.08 |18.67 |
|$22.83 |14.72 |21.71 |
|$27.46 |15.18 |15.34 |
|$18.55 |13.46 |16.19 |
|$23.77 |15.53 |11.85 |
|$19.35 |14.78 |13.76 |
|$27.01 |14.21 |18.26 |
|$20.66 |13.91 |19.23 |
|$22.27 |12.50 |15.02 |
|$19.65 |14.90 |15.70 |
|$21.71 |13.43 |12.25 |
|$23.25 |9.78 |14.18 |
|$26.26 |8.46 |13.88 |
|$20.07 |16.43 |11.28 |