An Information Management - Essay Example

?Section 2 a. There is wrong with the because it does not fit to how exactly the study was analysed in order to come up with the necessary result. The study did not mention on how the data was analysed and it seems by using the title “Marriage creates Happiness” is jumping right into a hasty conclusion. On the other hand, the data that was used or the samples in the study may not be representative or sufficient enough to represent the entire population of married couples in the world. Sample size and selection are primary considerations in order to come up with findings that are reliable (Anderson, 2004). b. Statistical methods are scientifically proven and they are widely used tools in order to describe relationships in data that are collected from research. Statistics is the art and science of collecting, presenting, analysing, and interpreting data (Broto, 2007; Medhi, 1992; Kazmier and Kazmier, 2009; Gravetter and Wallnu, 2008). Therefore, statistical methods are sufficient tools which are enough to describe relationships in data that are collected from research. Furthermore, they help researcher avoid committing mistakes and organise the entire research process (Bryman and Bell, 2007; Smith et al., 2002; Remenyi et al., 1998; Saunders et al., 2009; Howell, 2009; Armitage et al., 2002). c. For structured interview the proponent tries to conduct it for the purpose of getting data in a statistical survey. This is to ensure that all the questions are sufficiently answered and this is not only based on giving questionnaires for the respondents to answer. In this approach, the researcher or proponent of the study tries to organise the questions in detail for the respondents to answer and also in order to ensure sufficient and efficient data for statistical survey (Wilson and Sapsford, 2006). For example, a structured interview might be applied in a phone interview because of limited time involved. It is also in this set up that there is a need to substantially formulate concise but specific questions that are easy to understand or conceptualise by the respondents. The semi-structured interview on the other hand is unlike the structured interview because in the course of the interview, there are probably other questions that will be brought out aside from what are ordinary written on the questioner (Wilson and Sapsford, 2006). Mostly this type of interviews is conducted in social science studies. For example, this type of interview may be face to face where there is sufficient time for follow up questions. Such of this type of interview may be in line with studying current trend of the economy for instance in which the main respondents are persons of high calibrated information about the subject matter at hand. Furthermore, respondents may be those who have only small time for interviews yet with substantial knowledge on the subject matter. They can be managers, teachers or CEOs. The structured interview may be observed on television where some hosts have already prepared outlined questions. However, in order to make the conversations interactive some follow up questions are normally implemented in the discussion. The unstructured interviews on the other hand are entirely different from a structured interview because questions that will be brought up will depend entirely on the capacity of the respondents to answer the questions. Mostly, this type of interview is trying to evaluate the respondent’s intellectual capacity, beliefs or understanding on certain issues and questions will entirely depend on the evaluation made (Wilson and Sapsford, 2006). This type of interview is seldom made in phone interview because this will take time but most of this type in gathering data is conducted in the field of sociology. For example, this interview may be conducted in mental hospitals where patients’ behaviours need to be analysed. In all of these interviews, it can be observed that the data that will be obtained may be quantitative and qualitative depending on the type of question employed. d. Researchers try to engage in statistical hypothesis testing because testing inferences is a way of analysing and interpreting data (Buglear, 2004). In fact, the inferential statistics is an area in statistics which gives too much detail on analysis and interpretation of data. Statistical hypothesis testing is also a way to find if the analysed data have statistical significance. When the test is significant in testing inferences, this means that the null hypothesis is rejected and the alternative hypothesis is accepted. The converse is for testing inference with no significant result. e. Type I error is incurred when the null hypothesis is true but it is rejected and on type II error is incurred when the null hypothesis is accepted when in fact it is false (Calder, 1996). f. Reliability and validity are both common terms used in statistical analysis. A sample from a population must be reliable in order to come up with valid observation. For instance, in Plymouth Business School all female students can be randomly chosen in order to know there stand on abortion. However, it would be closely and time consuming to ask them all. One of the most common statistical strategies conducted in order to come up with information without using too many samples is to use 10 percent of female students who are to be selected at random. This ensures reliability and validity of the research process because one may be confident enough to generalise result based on choosing the reliable number of samples that can be representative enough of a certain population. Thus, part of reliability and validity process is to know the exact number of population so as to come up with reliable number of samples that will make the analysis valid at the end of the research process. Samples therefore are primary considerations in order to investigate the real information with utmost reliability (Anderson, 2004). This specific reliability will make the investigation valid. However, there is another way to come up with valid results and that is to ensure the use of scientific tools in the analysis. Statistical method is one. For many years, statistical methods have been reliably used to extract information from data set. This shows that statistical methods have been reliably used in the research process. g.i. The dependent variable for this model is risk measure and the independent or predictor variables are average number of cigarettes per day, age and blood pressure. This model is statistically significant because alpha which is 0.0 is less than 0.05 or 0.01. Conclusion about the inference is significant when p-values or alpha is less than the set level of confidence interval (Sternstein, 1994). Thus, at 95% level of confidence, the model is significant. On the other hand, at 99% level of confidence, the model is significant as well. g.ii. The dependent variable draws a regression line with data points. Based on the data, the R square is 0.869 which means that 86.9% of the data points are approximated exactly by the regression line. This only proves that there is higher information that can be explained by the dependent variable when it comes to forecasting behaviour. g.iii. Each independent variable is significant at 95% level of confidence because of their alpha number less that 0.05. However, all except cigarette per day is not significant at 99% level of confidence because its alpha is 0.014 which is greater than 0.01. This only implies that all independent variables have significantly contributed information for the dependent variable at 95% level of confidence. However, it is only ‘cigarettes per day’ which does not give so much significant information for the dependent variable at 95% level of confidence. Section 3 a. ANOVA analysis or Analysis of Variance is a way to compare the means of two or more groups of independent samples to find out if they have substantial variations (Karris, 2003; Hamilton, 2008; Delwiche and Slaughter, 2008; Kleinbaum et al., 2007). There are different kinds of analysis of variance and these include one-way analysis of variance, two-way analysis of variance, and three-way analysis of variance. The one-way analysis is suggested when only one variable is involved. The two-way analysis is suggested when there are two variables involved, consisting of column and row variables. Thus, the researcher would like to find out if there is any significant difference between and among columns and rows. The good thing about analysis of variance is that there is a good way to infer relationships existing between and among chosen data. For instance, as stated in the problem of Alicia, b. In general, I would suggest to Alicia that she needs to particularly look in general the confidence level involved and the statistical figure included. For instance, the first ANOVA table suggests that at 95% level of significance, there is a significant difference of the average number of times the customer eats out in a month and the food rating. In fact, the alpha obtained is 0.044 which is relatively smaller than 0.05. This means that the null hypothesis needs to be rejected considering that there is a considerable variations in the average number of times the customer eats out in a month and their food rating at 95% level of confidence. There is also significant difference between the eating out categories between less than 5 times and more than 10 times and the food rating. For Alicia, this information matters a lot and her main concern is to find which of these groups of customers have the promising consumption in a month based on their food rating and eating out paterns. Knowing this information, Alicia will be able to come up with decisions that will continue enhance her services so as not only to promote further to her promising customers but also to others who seldom eat out in a month. Furthermore, the obtained information would mean that these customers need to be closely determined in order to find out which among of the two categories of eating out patterns have the highest rating and the lowest rating compared to the other. This information gives Alicia a preliminary understanding which customers have considered excellent ratings for restaurants, cafes or pubs. Thus, her focus should be on customers with eating out pattern of less that 5 times and more than 10 times. In understanding this case, those customers with less than 5 times eating out pattern are most likely to have low personal food rating on restaurants, cafes or pubs considering the nature of their eating out patterns. These customers constitute the 21% of the overall samples. This further means that there is a considerable need to focus on them since they can also give out certain market share for Alicia. On the other hand, Alicia was also able to gather the approximate annual income of her customers. There is a good reason to actually know this since it is intuitive that customers’ spending is associated with their income. Thus, it is important on the side of Alicia to know if there is a significant difference of her customers’ income and their food rating in particular. From the second ANOVA table, the alpha or p-value is 0.179. This value is relatively greater than 0.05. Thus, at 95% level of confidence, the null hypothesis is accepted. This means that there is no significant difference in the approximate annual income of Alicia’s customers and their food rating. Furthermore, by comparing the actual income of respondents from each category, there was no significant difference in food rating. Alicia can further deduce additional information from this that it is not actually the customers’ income that affects their average eating out of customers in a month particularly on their food rating. c. Considering that there is no significant variation in the approximate income of customers in a month and their food rating, Alicia can eventually focus on the idea that somehow there must be a necessary evaluative measure she can do in order to improve further her services. This is due to the fact that there might be substantial reason to be improved with regards to her services when it comes to customer’s eating out pattern and their food rating. Some of these customers might perceive eating out in the restaurant unnecessary or not worth for their money. Thus, there is a need for Alicia to implement innovative measures in order to influence customers’ behaviors when it comes to eating in restaurants, cafes and pubs. Alicia must understand by now that approximate annual income cannot be a significant predictor of food rating on restaurants, cafes and pubs. Instead, it is important to give considerable attention on customers’ food rating and eating out pattern. The goal now is to influence and create needs for those customers who most likely have an eating out pattern with unsatisfactory food rating on restaurant, cafes and pubs. References Anderson, V. (2004) Research methods in human resource management. Great Britain: CIDP Publishing. Armitage, P., Berry, G., and Matthews, J. N. S. (2002) Statistical methods in medical research. United Kingdom: Wiley-Blackwell. Bryman, A. and Bell, E. (2007) Business Research Methods. 2nd ed. Oxford: Oxford University Press. Broto, A. S. (2007) Inferential Statistics. Manila: National Book Store. Buglear, J. (2004) Quantitative Methods for Business: A to Z of QM. Oxford: Elseier Butterworth-Heinemann. Calder, J. (1996) ‘Statistical Techniques’, in R. Sapsford and V. Jupp (eds). Data Collection and Analysis. London: Sage. Delwiche, L. D., and Slaugher, S. J. (2008) The little SAS book: a primer. 4th ed. USA: Cengage Learning. Gravetter, F. J., and Wallnu, L. B. (2008) Statistics for the Behavioral Sciences. 8th ed. USA: Cengage Learning. Hamilton, L. C. (2008) Statistics with Stata: updated for version 10. 7th ed. Canada: Cengage Learning. Howell, D. C. (2009) Statistics Methods for Psychology. 7th ed. USA: Cengage Learning. Karris, S. T. (2003) Mathematics for business, science, and technology: with MATLAB and spreadsheet applications. USA: Orchard Publications. Kleinbaum, D. G., Kupper, L. L., Muller, K. E. (2007) Applied regression analysis and other multivariable methods. USA: Cengage Learning. Kazmier, L. J., and Lazmier, L. (2009) Schaum’s Outline of Business Statistics. 4th ed. USA: McGraw-Hill Professional. Medhi, J. (1992) Statistical methods: an introductory text. New Delhi: New Age International. Remenyi, D., Williams, B., Money, A., and Swartz, E. (1998) Doing Research in Business and Management: an introduction to process and method. London: SAGE. Saunders, M., Lewis, P. and Thornhill, A. (2009) Research Methods for Business Students. 5th ed. London: FT Pitman. Smith, M. E., Thorpe, R., and Lowe, A. (2002) Management Research: an introduction. 2nd ed. London: SAGE. Sternstein, M. (1994) Statistics. USA: Barron’s Educational Series. Wilson, M., and Sapsford, R. (2006) ‘Asking Questions’, in R. Sapsford and V. Jupp (eds). Data Collection and Analysis. 2nd ed. London: Sage. Read More