Applied Statistics and Forecasting: Relationship between Salary and Casual Factors
Verified
Added on  2023/06/12
|15
|3132
|329
AI Summary
This study analyzes the relationship between salary and casual factors using inferential tools such as Durbin-Watson test, residual analysis, collinearity, and error distribution. The study also examines how innovation in UK companies has changed through its casual factors. The research type is quantitative, and SPSS software is used for data analysis.
Contribute Materials
Your contribution can guide someone’s learning journey. Share your
documents today.
BE279 Applied Statistics and Forecasting
Secure Best Marks with AI Grader
Need help grading? Try our AI Grader for instant feedback on your assignments.
TABLE OF CONTENTS INTRODUCTION.........................................................................................................................2 METHOOLOGY AND DATA.....................................................................................................2 ANALYSIS AND RESULTS........................................................................................................3 Task 1..........................................................................................................................................3 Task 2..........................................................................................................................................7 DISCUSSION...............................................................................................................................11 CONCLUSION............................................................................................................................12 REFERENCES............................................................................................................................14
INTRODUCTION Statistical forecasting is mainly implies that the use of statistics based upon the historical data of a project so that it will be easy to determine the same in future. This is mainly used in the quantitative data in order to determine the interdependence between the variables. The present study is also helpful to develop a deep understanding pertaining to statistics and tool used to determine the answers. For that, there are two topics on which the entire study is based. such that under first, study will determine the relationship between salary and other casual factors by using an inferential tool. Thus, to check the autocorreltion, different test will be applied which include Durbin-Watson test, residual analysis), collinearity (VIF), and error distribution. Moreover, under topic 2, the study will examine how an innovation developed by the companies has change through its casual factors and by using different inferential tool, the present study will explain the results. METHOOLOGY AND DATA Research type:Quantitative research type has been adopted for the present study in which numbers and figures has been considered in order to derive a valid outcome (Nayak and Singh, 2021). In order to analyse the association between the variables, only quantitative research type will be beneficial. Research approach and philosophy:In accordance with the research type, deductive approach and interpretivism philosophy has been adopted that assist to examine the relationship between salary and casual factors effectively (Snyder, 2019). This in turn assist to determine the valid outcome and answer the research questions as well. Data collection:The entire study is based upon secondary data collection methods in which responses have been gathered from previously conducted research so that the relationship can be identified. In this, 787 people has been interviewed and this is further supported by different inferential tool so that effective outcome can be generated (Mohajan, 2018). Along with this, under discussion chapter, the sources has been further selected that helps to support the results so that effective outcome can be generated.
Data analysis:In order to analyse the data effectually, SPSS software has been used which help to ascertain the hypothesis and formulate the results (Pandey and Pandey, 2021). In this, regression and anova as an inferential tool will be used to determine an association and further with the help of table and graphs the report will present the data in an effective manner. ANALYSIS AND RESULTS Task 1 Null hypothesis (H0): There is no significant relationship between casual factors (publication, position and university) upon likelihood of faculty salary. Alternative hypothesis (H1): There is a significant relationship between casual factors (publication, position and university) upon likelihood of faculty salary. Model Summaryb ModelRR SquareAdjusted R Square Std. Error of the Estimate Durbin-Watson 1.659a.434.42820677.9101.396 a. Predictors: (Constant), prof, osu, female, pubindx, assist, assoc b. Dependent Variable: salary ANOVAa ModelSum of SquaresdfMean SquareFSig. 1 Regression185458035781. 828630909672630.3 0572.290.000b Residual241580416113. 242565427575957.723 Total427038451895. 071571 a. Dependent Variable: salary b. Predictors: (Constant), prof, osu, female, pubindx, assist, assoc Coefficientsa
Paraphrase This Document
Need a fresh take? Get an instant paraphrase of this document with our AI Paraphraser
ModelUnstandardized Coefficients Standardized Coefficients tSig.Collinearity Statistics BStd. ErrorBetaToleranceVIF 1 (Constant)64884.7886643.6159.766.000 assist- 14300.3757165.398-.152-1.996.046.1735.775 assoc-9498.5736878.987-.135-1.381.168.1059.557 prof15387.3426668.421.2572.307.021.08112.383 female2648.3413283.388.027.807.420.9191.088 osu3606.3153038.545.0381.187.236.9801.021 pubindx255.48622.857.37911.178.000.8691.151 a. Dependent Variable: salary Collinearity Diagnosticsa ModelDimensio n Eigenvalu e Conditio n Index Variance Proportions (Constant ) assistasso c pro f femal e os u pubind x 1 12.8381.000.00.00.00.00.01.02.04 21.2861.485.00.02.01.00.23.01.03 31.0141.673.00.08.03.00.01.05.00 4.8451.832.00.02.00.00.02.90.00 5.7022.011.00.03.03.00.66.00.00 6.3063.045.01.00.00.01.07.00.93 7.00918.121.99.85.92.98.00.01.00 a. Dependent Variable: salary Residuals Statisticsa MinimumMaximumMeanStd. DeviationN Predicted Value50584.41143862.5283674.7018022.074572 Residual-60905.223110653.781.00020568.982572 Std. Predicted Value-1.8363.340.0001.000572 Std. Residual-2.9455.351.000.995572 a. Dependent Variable: salary Charts
Interpretation:Through the model summary table, it has been analysed that the there is a moderate relationship between salary and casual factors which in turn reflected that there is a change identified over salary when a minor fluctuation identified over casual factors. Moreover, as per the r square value it has been also identified that there is only 43% change in the salary of the selected respondents. This in turn entails that there is a change examined over dependent variable due to fluctuate in independent variable. As per the anova table, it has been reflected that there is a significant difference between the salary and other casual factors because the value of p is 0.00 which is lower than 0.05 and this in turn reflected that alternative hypothesis is accepted. That is why, it can be stated that due to change in publication, states and gender, the salary will be fluctuated.
Secure Best Marks with AI Grader
Need help grading? Try our AI Grader for instant feedback on your assignments.
Moreover, with the help of durbin Watson test (value is 1.3), it has been identified that there is positive autocorrelation between the variable and that is why, the variables i.e. salary and casual factors have a direct relationship with others. As per the scatter plot and residual statistics, it has been identified that when plotted against the predictor variable, the variability is constant across different values of predictor variable. That is why it can be stated that the observed value did not be varies more from predicted one and that is why, it helps to generate a better result. According to collinearity statistics under coefficient table, it has been identified that the value obtained VIF value is lies in between 0 to 10 in all cases except professor. This in turn shows that only in professor case, the collinearity issue has identified which need to be removed in order to generate a better outcome. The error of distribution can be determined under a graph named as observed cumulative probability in which little circles follow the normality line and that is why, it can be stated that there is no drastic deviation and the inter-dependence of the variables may be depend upon each other. Task 2 Factor analysis KMO and Bartlett's Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy..741 Bartlett's Test of Sphericity Approx. Chi-Square3589.475 df10 Sig..000 In accordance with the above table of KMO and Bartlett’s test it has been identified that the sample size is sufficient for factor analysis and the significant value is less than 0.05 which in turn shows that sample is too small. However, on the another side KMO value is 0.74 which reflected that the sample size is sufficient. On the other side, as per Bartlett’s test of sphercity which entails that the scholar have enough number of correlation between the variable for factor
analysis because the value of less than the standard criteria. Hence, it can be reflected that the variables have enough correlation between the variables in order to analyse the data effectively. One way anova H0: There is no statistical difference between the innovation and size or cooperation H1: There is a statistical difference between the innovation and size or cooperation ANOVA Sum of Squares dfMean SquareFSig. size of company (staff) Between Groups562134.9993118133.3872.080.000 Within Groups42627302.56448908717.240 Total43189437.5634921 Cooperation Between Groups208.855316.73743.042.000 Within Groups765.4224890.157 Total974.2774921 Interpretation:In accordance with the above table, it has been identified that there is direct relationship between innovation in UK companies and casual factors such that the alternative hypothesis is accepted over the other. The reason for selecting this hypothesis is due to the value is lower than 0.05 and this in turn shows that the alternative hypothesis is accepted. Apart from this, as per the model summary, there is low association between the variable and that is why, it can be stated that there is minor change noted over innovation within companies. Moreover, there is only minor change identified over innovation when the independent variable changes which reflected that though the changes has identified but there is an impact identified over both variable. Clsuter analysis
Initial Cluster Centers Cluster 123456789101112131415 size of compa ny (staff) 311434121342432 Cooper ation100101100011001 REGR factor score 1 for analysi s 1 .40 523 -.83 769 1.61 216 1.61 216 3.03 895 -.83 769 .58 910 .40 523 3.03 895 1.48 589 3.03 895 -.83 769 .40 523 -.83 769 2.24 122 Iteration Historya IterationChange in Cluster Centers 123456789101112131415 1.181.643.618.181.455.588.608.237.000.245.028.473.119.250.439 2.120.014.279.119.170.000.265.135.839.058.196.085.092.000.297 3.208.006.068.047.293.000.164.027.449.427.018.068.088.000.191 4.132.000.018.005.084.003.032.000.102.063.018.000.000.007.199 5.015.000.000.000.014.023.016.000.213.000.000.008.000.000.064 6.004.000.000.000.000.000.006.000.190.000.000.000.000.000.090 7.000.000.000.000.000.000.028.000.000.000.000.000.000.000.027 8.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000 a. Convergence achieved due to no or small change in cluster centers. The maximum absolute coordinate change for any center is .000. The current iteration is 8. The minimum distance between initial centers is 1.414. Final Cluster Centers Cluster 123456789101112131415
Paraphrase This Document
Need a fresh take? Get an instant paraphrase of this document with our AI Paraphraser
size of compa ny (staff) 322434222342433 Cooper ation100100101011001 REGR factor score 1 for analysi s 1 .21 653 -.81 170 1.41 498 1.65 127 2.64 904 -.73 102 .75 253 .01 681 2.54 909 .70 937 3.00 655 -.38 827 .10 593 -.59 431 1.59 275 Number of Cases in each Cluster Cluster 1404.000 21389.000 3155.000 4204.000 596.000 6265.000 7160.000 8605.000 958.000 10274.000 1156.000 12210.000 13137.000 14731.000 15178.000 Valid4922.000 Missing4505.000
Interpretation:As per the above final cluster, it has been interpreted that the cluster 1 is very far from the other profile and the same has been identified over the other cluster. This in turn reflected that there is a need to work on the innovation so that the size of the companies cannot be affected and as a result, the cluster will be close towards each by sharing a bond. In accordance with the cluster 2, it has been identified that it has a highest cluster and thus it assist to determine the views in a group. DISCUSSION In accordance with the topic 1, it has been identified that there is a direct relationship between salary and other factors. This has also supported by Rudakov and Roshchin (2019) that
the salary of an individual is mainly affected by publication, position, universities and state. Such that the state is located in poor country then the salary of an individual will be low and publishing also enforce an academic’s reputation as a scholar and offer many benefits as well (Cullen and Perez-Truglia, 2022). That is why, it shows that when the casual factors changes, there will be change identified over salary and that is why, it affects the individual salary directly. On the other side, it has been identifiedunder topic 2 thatcompany’ssize and cooperation always have a direct impact over likelihood of innovation which in turn prove alternative hypothesis under one way anova. In the opinion of Lin and et.al., (2019) as compared to small firms, larger firms are more likely to control over the resources which is necessary for the innovation and this also include capital as well. That is why, it can be stated that productivity of small firms is lower and that is why, there is a huge difference between the mean of all size group among companies. Moreover, there are different resources which also have a direct impact over the innovation and this might be affected the innovation of UK companies. Andries and Stephan (2019) argued that small business are considered as a prime for innovation and that is why, the number of small business in UK are easily increases from last many years. Overall, through the above analysis, it has been identified that there when the changes identified over the casual factors then it affected the salary factor of the employees as well. As companyoffersalarybyconsideringmanyfactorswhichincludeeconomy,experience, publication, gender etc. That is why, according to regression analysis, alternative hypothesis has accepted over other. This shows that employers need to include further actions as well in order to generate a better outcome (Li, Liao and Albitar, 2020). However, the likelihood of innovative is also changes by considering the other factors like most of the companies have effective employees which in turn assist to improve the performance and they all involve within a company that helps to make the company more innovative and generate a better outcome as well. This in turn shows that the size of the company is actually affected the likelihood of innovation in which company actually affected the overall performance in positive manner.
Secure Best Marks with AI Grader
Need help grading? Try our AI Grader for instant feedback on your assignments.
CONCLUSION By summing up above report it has been concluded that statistics forecasting will be more beneficial for the companies which provide accurate results. Through topic 1 it has been summarizedthatthereisamoderateassociationbetweenprofessionalcharacteristicand publication issues in term of salary faculty profile. Thus, alternative hypothesis has been accepted over the other which in turn shows that there is a fluctuation identified over salary when other independent variable changes. Moreover, the Durbin Watson test also indicated that there is a positive autocorrelation between the variable and this in turn helps to generate a better outcome. As per the topic 2, it has been concluded that innovation is actually affected by the size and cooperation agreements in all the selected companies. Thus, innovation has vary from all the different sizes of the companies and that is why the value of p is lower than standard criteria which in turn accept alternative hypothesis. In accordance with the cluster analysis, all the cluster is very far from the profile and this in turn shows that changes are actually affected.
REFERENCES Books and Journals Andries, P. and Stephan, U., 2019. Environmental innovation and firm performance: How firm size and motives matter.Sustainability.11(13). p.3585. Cullen, Z. and Perez-Truglia, R., 2022. How much does your boss make? the effects of salary comparisons.Journal of Political Economy.130(3). pp.766-822. Li, Z., Liao, G. and Albitar, K., 2020. Does corporate environmental responsibility engagement affect firm value? The mediating role of corporate innovation.Business Strategy and the Environment.29(3). pp.1045-1055. Lin, W.L. and et.al., 2019. Does firm size matter? Evidence on the impact of the green innovation strategy on corporate financial performance in the automotive sector.Journal of Cleaner Production.229. pp.974-988. Mohajan,H.K.,2018.Qualitativeresearchmethodologyinsocialsciencesandrelated subjects.Journal of Economic Development, Environment and People.7(1). pp.23-48. Nayak,J.K.andSingh,P.,2021.Fundamentalsofresearchmethodologyproblemsand prospects. SSDN Publishers & Distributors. Pandey, P. and Pandey, M.M., 2021.Research methodology tools and techniques. Bridge Center. Rudakov, V. and Roshchin, S., 2019. The impact of student academic achievement on graduate salaries: the case of a leading Russian university.Journal of Education and Work.32(2). pp.156-180. Snyder,H.,2019.Literaturereviewasaresearchmethodology:Anoverviewand guidelines.Journal of business research.104. pp.333-339.