Microsoft company overview and analysis

MIcrosoft

Business Research Issues

Any business that wishes to remain successful must conduct some level of ongoing research into various aspects of their business operations and effects. Microsoft is certainly no exception to this rule, and research into its level of customer satisfaction can suggest certain pitfalls that the company is endanger of falling into as well as providing clear and concrete methods by which the company can improve its sales and profitability. In order for such research to be effective, however, there must be proper controls on the validity and reliability of the data being collected and the methods of analyzing and interpreting such data. The following pages will examine specific ways by which the validity and reliability of data collected through customer surveys can be assured, and how that data can be put to direct and effective use within the Microsoft Corporation't improve its customer satisfaction levels.

Statistical analysis will necessarily be a major part of assessing and interpreting the information collected via these surveys, and there are several established methodologies for achieving reliability and validity in such interpretations, as long as the data is collected properly and is pertinent to the research questions beings asked (Burns & Burns 2008). SPSS software is the most commonly used statistical tool for statistical analysis, and will be employed to analyze this research; the non-quantified raw data will have to be converted into quantities before this tool can be utilized, however (Burns & Burns 2008). Following this quantification, possibly using a binary Likert scale for the ordinal ranking elements and using other methods for the nominal data, analysis will be possible (Herbst & Coldwell 2004).

There are certainly some challenges that can be raised in regards to the validity and reliability of the data collected through this survey. First, it is difficult to predict what the response rate will be, and it will be harder still to ensure that respondents are actually owners/users of the Microsoft products being examined -- certain controls will have to be set up to ensure greater reliability (Landstrom 2009). In addition, the quantifying of qualitative data always creates certain validity issues, and these are somewhat exacerbated by the lack of any open-ended space for explanation or further reflection on the part of the respondent, which can help to code and correctly apply responses in the manner intended y the researchers in their posing of research questions and creation of the survey instrument (Burns & Burns 2008; Herbst & Coldwell 2004). The length of the survey and the simplicity of the responses, while increasing the reliability of the data somewhat, may also harm its validity.

In order to minimize the challenges and potential problems that exist with this research methodology, there are several steps and/or adjustments to practice that should be considered. When it comes to the instrument itself, including open-ended spaces for respondents to give a more detailed explanation of their responses might help in the later classification and quantification of responses, and will certainly provide valuable qualitative information regarding customer satisfaction. This very slight adjustment to the survey instrument might also promote more careful consideration and a greater level honesty among the respondents, making the directly0quantifiable answers more valid as a result (Herbst & Coldwell 2004).

Other steps that should be considered to ensure greater validity and reliability include taking appropriate measures to ensure that respondents are indeed users of the Microsoft products being researched, where applicable, and other demographic and computer use details should also be collected to determine if certain populations are more prone to specific perceptions or responses than others (Landstrom 2009). When conducting the statistical analysis of the collected data, this extra information can greatly assist in drawing correlations and comparisons of data groups, providing a more solid foundation from which to draw statistical inferences from the analyzed data (Landstrom 2009; Burns & Burns 2008). Other standard research practices, such as controlling for other variables that might be present and maintaining proper distance to avoid bias, should also be utilized.