The Artsy Corporation Essay Example For Students

The Artsy Corporation Essay They make a statement that gender would effect certain factors such as: pay, hiring, promotions and Other elements Which are involved with the career of Artsy employees. This paper will put together statistical information and analysis based on the data we retrieve in order to give Arterys lawyers a perspective of what the company truly looks like in order for them to be able to put together a defense. The data put together for us to analyze is based on the payroll of Arts 256 employees at one of their facilities. This specific data was selected by usage of simple random sampling (the sample represents the population, i. E. Every employees pay rate and working conditions). The data includes: an identification number (Daimler) that would permit us to identify the person by name or social security number, the persons sex (SEX) where a O denotes female and a I denotes a male, the persons job grade in 1986 (GRADE), hierarchy level at company the length Of time (in years) the person had been in that job grade as Of 12/31/86 (TING), and The persons weekly pay rate as of 12/31/86 (RATE). This being the most important point Of concern) In order to analyze this data statistically we will create multiple regression oodles. In this case, we will consider Pay Rate as a dependent variable and Gender, Job Grade and Time in Grade as independent variables. We will work through the data and try to find evidence to show whether or not gender plays a role in the company by influencing the salaries. On the other hand, if we find out that gender does not play a role in influencing the salary then we will try to find the other independent variable that does make a difference. Art l: Descriptive Statistics Defining Important Terms In this paper we will be making use of descriptive statistics which is basically a ay of collecting, summarizing, and analyzing our data in order to come to a conclusion. This is very valuable for us since it will allow us to analyze of large group of numbers such as the set of data we are considering. When analyzing each specific set of data we will be considering both the central tendency and variation of the specific variable we are discussing. Also in this paper we Will refer to the mean in specific parts, where the word basically refers to a measure of central tendency, it is the arithmetic average of the numbers in our data. In order for our reader to have a better understanding of mean and arithmetic average. E have given an example below: By using our data and the soft. are available to us, we were able to conclude that the average pay rate of females in Artsy Corporation was $833 per week, whereas the average pay rate of males in Artsy Corporation was $1128 per week. With this information we can have an idea that males have a higher salary (This is a quick example which does not take the other variables into consideration). The last important term that is necessary to clarify in order for a better understanding is: standard deviation. The standard deviation is a number that measure the spread tooth data in relation to the mean number. It gives us the scatter of the information in terms to percentage, giving us an idea of how close together or separate our dull data set is, Data Analysis of Pay Rate Now that we have a better understanding of the concepts we can start analyzing our data table. Our first variable to analyze is pay rate, since it is the one of our concern. The lowest pay rate is $579 per week, and our highest is $1552 per week, giving us an average pay rate of $931 per week. Our first standard deviation is $229 per week. Basically vat this means is that 68% of our data (1st standard deviation) is either $229 above or below our mean Of $931. Below we have put together a box-whisker plot of the pay rate in order to have a better idea Of the salaries at this corporation: Figure I In the plot above we are able to come to various conclusions. The first points we have to analyze are our quartiles: 25% of the employees have a weekly pay rate less than 5762. Of the employees have a weekly pay rate between $762 and $1073. 25% of the employees have a weekly pay rate above $1073 Our middle number here is $865, which means that of our employees make more than this, and 50% make less The little red box on the right of our ATA is an extreme outlier which basically signifies a point which is out of the context compared to other numbers. We assume that this number is the pay rate of the executive or manager of the branch Which is much higher than those Of the employees, therefore it should not be used to compare to others. By simply looking at this graph we are able to see that the lower 25% Of the salaries are much more agglomerated at the end and seem to all be in the same range. However, the salaries Of those in the highest Of the data are much more spread out since the line is much longer at this end. The lines at the end of the ox give a sense of how close together, or spread out our information is. In this case the higher salaries are more spread out. Data Analysis of Pay Rate vs.. Gender In the introduction we talked about the average salaries of men and women. We discussed how they were very much different even though we still had not taken into consideration the other variables given to us in the data set. Below we have created a box and whisker plot where the pay rate of the men and women of the company are put side by side in order to have a better idea to how they differ, Figure 2 Prom the data displayed in the plots above we can see that there is a big preference between male and female salaries. The bottom 25% of the male employees make approximately the same amount as the women in the intermediate portion (from 25%-75%, middle 50%). Since this is a sample which we Will base our Whole population on, then we can say that this information pertaining to this sample represents our companys numbers in general. Also, once again there are red squares to the right Of our plots. In this case there are 4 red boxes (2 are overlapping showing a darker tone of red), which show that the higher paid females in the company are seen as outliers when analyzing them statistically. Women that have a higher pay rate are not in the normal patterns of Artsy and therefore are statistical outliers compared to the rest of the company. Once again, since this is simply an initial data observation without taking into consideration our other variables we cannot make a conclusion, however; we can see that there is a tendency of higher salaries towards men, and we can infer that there is a possibility that gender does play a role in pay rates. Data Analysis of Grade We can now analyze the grade of our employees and how this affects the salary. The grade is simply the hierarchical position of our employees in the company here they are scaled from I to 8 (l being the lowest, and 8 being the highest). Below is a table of each of our employees in terms of grade. Figure 3 Looking at the graph, we can see that there is a similar distribution of employees at both the top management positions, and also the lower position. It seems to be very well distributed which takes away the idea that the higher the grade the fewer employees there are. Figure 4 Above in Figure 4 we have put together a scatter plot Of our employees pay rate and how it differs as they go up in their grade levels. Each circle represents one employee and their position symbolizes What level they are at, and What pay rate they are given. Also we have made a gender code where we are able to see the difference beet. en males and females. The red box (I) represents male employees, while the blue circle (O) represents the female employees. There is clearly a trend in this graph where the higher an employee is based on his grade level, the higher his salary will be. Supply And Demand EssayOnce again this shows the traditional hierarchy in companies, where higher level employees receive higher salaries. The RE of this equation is 823%. The R? Shown above together with the equation tells us how much to the variation in these pay rates are due to the variables which we have just described. This means that in this case, with an RE value of 823%, 823% of all the differences in pay rate in our set of data are explained with the use of our independent variables (gender, grade, time in grade), This tells us that statistically we can correctly prove there is a relation between the pay rate and our independent variables of the time. The last part to our statistical information above is the S. E (Standard Error of the Estimate), which is equal to 97,06_ The S. E. Displays the variation of our predictions. This number shows how our predictions might fluctuate, basically meaning that by using this equation to predict our employees pay rate eve can be off by  ±$go. Do per week Basically we can have a number that is either over or under the actual number by $97. 06. This tool is very important because it slims down our margin for error. Before in this paper we mentioned that our standard deviation for our pay rate avgas $229 per week. NOW With the regression model we have come to a lower error Of $97. 06. We have reduced our error percentages by about 58% with the regression model. NOW that we have already created our initial regression model it is necessary for us to create individual regression models for each of the variables in order to find out how much of the variation is caused by each of the independent variables. Our first individual regression model is using our gender variable. The model that will predict the pay rate with just gender as a predictor variable is: Pay Rate 833 295 Gender Coded As we have discussed in our initial model, since gender is either male or female being a qualitative variable, it is displayed as either 1 or 0. In this case we are able to see that on average, a males pay rate is per week higher than the pay rate of a female. In the previous regression model this number was much lower ($59. 6), however, here we are individually selecting variables and isolating them in order to find out how much each of the variables affect the pay rate. Since our RE here is 36. %, we can say that 36. % of the variation in pay rate may be explained simply by knowing the gender. The second individual regression model that we Will create is considering time within grade. The model that will predict the pay rate with just time in grade as a predictor is: Pay rate = 788+823 (Time in Grade) This model tells us that for every additional year Within a grade l evel the pay rate increases, on average, $82. 3 per week. However. Since here we see an RE which is a low 29%, we can come to the conclusion that the time a person has worked within a grade level is not significant in terms of their pay rate. We believe that his happens because the time within grade does not account for the experience the person has in the company overall. A person in grade 8 might have 10 years in the company, however he might have just been promoted to grade 8 therefore he has a low time within grade. On the other hand there might be someone in grade 1 for the past 5 years. Obviously their pay rate is cannot be based on time within grade even though it has some affect to the overall model, However, with this percentage to we consider this to be insufficient proof of variation. Our last individual regression model which will be created is considering the grade bevel. Once again we have made changes to this variable where we separated out the grade level variables in order to create a specific variable for each grade level. We have created 8 variables with a 1 and 0 possibility, 1 being in that grade level and O being that they are not in that specific grade level. Below is our individual grade regression model: Rate 671 694 Grade_8 501 Grade_7 + 385 Grade_6 226 Grade_S 161 Grade 4 +161 Grade 3+54. Grade 2 S=ASSESSES/week Above eve have displayed how the pay rate increases in each grade vivid respect to grade level I(the initial grade level at the company). What this basically means s that grade level 8 employees on average make $694 per week more than grade level I employees. At the same time grade level 4 employees make $161 per week more than grade level 1 employees. All the grades in the model above are being compared to the lowest grade level possible ( grade level I) since it is not on the regression model, and we can relate all numbers back to it. Now in terms of the RE, we see that there is a very large number in comparison to all the other individual models. Basically 81. 8% of the pay rate can be explained simply by knowing the employees grade level. The lawyers defending Artsy can definitely SE this information since they can say that 82% of the time the pay rate is defined due to the employees level inside the company. It can also be compared to the individual gender model to show how the 82% is much more significant than the 36. 9% we found before. After running the initial regression model together with all of the individual ones, we have come to develop a new regression model where we will use our grades separately as we have shown above in order to see if we increase our RE and reduce our Standard Error. The new regression model is displayed below: Rate 632 46. 9 Gender Coded 26,9 Delegated 60. Grade_2 Carried_3 168 Grade 210 Grade 5 + 356 Grade 6+ 434 Grade 7 * 613 Grade 29/week Once again this model shows how the pay rate changes with respect to each individual variable. In the gender variable it tells us that on average men make $46. 9 per week more than women at Artsy Corporation. Also in terms of the grade levels we need to remember that each grade level coefficient is being shown with respect to the lowest grade level at the company (grade level 1). All the ways Of interpreting this model Will remain the same as the ones Which we have shown before. Since this model has shown us a larger RE of 85. 4% imparted to our 823% from our initial model, we will continue to use the new model since it displays where the variation comes from at a more accurate level. Also another factor that shows us that our new regression model is better from our old one is our S. E. The standard error in our initial model was $97. 06 per week, whereas our new regression model has an S. E of $89. 29 per week. We have reduced our error by SO,77 or Once again this is a good sign since it gives the lawyers a more accurate display of information and reducing our error in order for them to have a better more solid defense when making the case against ender discrimination. Before we can start using our regression model we need to check and see if all of revisable can be present in our model. In order for variables to be accepted in regression models they need to be in two specific conditions: linearity and equal variance. We will check these two conditions using a normal plot of residuals and residuals versus fits plot. We would first want you to understand what linearity and equal variance means, Equal variance means that the variability in pay rates is the same regardless of the independent variables having either high or low values Linearity is to see when the pay rates vary erectly in proportion to our independent variables (gender, time within grade and grade level).