Chapter 3 Research Methodology
3.1 Introduction of Company A in Study
Company A is one of the leading window covering manufacturers in Asia, exporting a full line of mini blinds, roll up blinds, vertical blinds, shutters, roman shades and convolute rollers－all manufactured to their customers specifications as part of their private label program. Corporate headquarters and show room are located in Taipei close to the Taiwan World Trade Center and easily accessible to visiting buyers. Products are manufactured at our two factories in China. These facilities have more than 60,000 sq. meters of manufacturing space, equipped with up-to-date equipment and staffed by management with extensive experience in plastic technology, fabrication and assembly. Both are completely self-sufficient facilities directed from the company’s Taipei headquarters. All PVC products are manufactured in facility No. 1, while textile and woven wood products are produced in facility No. 2.
A third facility will be located in North Vietnam which is one hour away from the Hanoi Airport with more than 90,000 sq. meters of manufacturing space. This company is strong and continues to grow in size.
Company A is dedicated to continuous development of new products and improvements that respond to market demand. A continuous flow of new product is the life for a company to remain competitive in its industry. (Barczak, 1995) Company A stays in close contact with markets worldwide through their US and Canada liaison offices. This has contributed to company A leading in development of
“Insulated Cellular” venetian blind slats; “Clear-View” room darkening blinds; a new
“Luxer” gloss finish to give practical vinyl an aluminum look; and patented
“Child-Safe” cordless systems for both blinds and roman shades. Other quality innovations are under development as part of a continuing program to supplement its basic assortment of high volume promotional products.
Company A has been in this industry for over 30 years and stays in the top three major manufacturers in China and is a current vendor/supplier for major retailers in USA including but not limited to, Home Depot, Lowes, JC Penney, etc. Not only has their sale growth in the past 15 years doubled to over 75 million dollars, their number
30
of employees has been doubled with more than 1300 people today. Their gross profit has increased as much as 20% compared to 15 years ago. The key to its success lays in their fast development of new product innovation and acute sense of adaptation prior to the market change. A firm deals with the market intelligence to facilitate the customer needs and generate appropriate response to such needs is the key to success.
(Borges, Hoppen, & Luce, 2009) As the technology advances throughout the years, Company A has been implementing a new management system from the inspiration of Toyota, the automaker; new automated machines for the manufacturing process; and new training programs for its employees to become more diverse and trained with multiple skills rather than single task work.
However, both labor and material costs have increased over the years and the profit margin has been reduced. It is important to micromanage the business by more detailed analysis in its sales.
3.2 Product Portfolio of Company A
It is estimated that there are over 100 million units of hard window coverings shipped from China to United States and Company A has about 10% of its market share. See below pie chart of an overall view.
Figure 18: Product Type Category in units of Company A.
31
From figure 18, the pie chart shows that there is a significant quantity of mini blinds manufactured by Company A, it is over 50% which is more than half of what Company A has manufactured. The second highest product is the fauxwood blind which is another type of venetian blind but premium. The roller shade is in third place.
However, the figures would be different in dollars, it is a different story.
Figure 19: Product Type Category in dollars of Company A.
In figure 19, the pie chart shows products in dollars of Company A, fauxwood Blind has a higher ratio than the mini blinds. Fauxwood blind is indeed a premium product which costs more to manufacture. Mini blind is a commodity product which is sold by bulk size. Thus dollar value of a mini blind is much lower than the fauxwood blind. Another very extreme product is the Shutters, it consists only 1% of units but 11% of dollars. That’s 11 times difference between dollars and units. Shutter is a very high premium product with extremely high costs to manufacture. For the purpose of this research, it is important to categorize them into different types of categories by their light control. It is difficult to analyze them by product type categories as it has no relationship with our three factors.
It is important to review these products into three different product categories as specified in table 1 in both sales units and dollars. Those three product categories are Light Filtering, Room Darkening and Light Blocking. For the purpose of this research, comparison can be made between the sales of stock sizes and the online sales. Here is
32
the layout of all three based on the units and dollars.
Figure 20: Product Category in Units (top) and Dollars (bottom) of Company A.
The light filtering category has over 50% while light blocking only has 31% in units, but in dollars, the light blocking has over 50% and light filtering has 28%. The data clearly shows that light blocking is more expensive than the light filtering in cost.
33
It would be interesting to compare this with the online sales. See figure 21 for the online sales.
Figure 21: Product Category in Online Sales in Units (top) and Dollars (bottom) of Company A.
The data of online sales is telling a different story than the stock sizes. This data
34
clearly shows the online shopping focuses on more expensive and light blocking window coverings. Because of these differences between stock size sales and online sales, this research looks into the three factors that are affecting the online business.
These factors are environment, economics and demographics in the United States.
Table 2: Key Factors and Their Elements with Brief Descriptions
The above table is a list of this research looking at the six elements that may which means that the products are manufactured and shipped to vendor’s warehouse in United States. These products are sold through the retailers’ online business; and then courier delivered directly to the consumer. The original data consist of sales units and dollars of each state and filed by different retailers. The data also consist of product returns which were the result of incorrect orders or damaged during shipping and are removed from the data. Finally, these data of different retailers are then pooled together in one large organized data.
However, as there are many different varieties of products in the pooled data, products are then categorized into three types by sharing the same common properties of light control. Please refer to table 1 in chapter 1.
Geographic data are collected from the original data and further analysis is done by different regions in United States. United States Census Bureau defines four statistical regions, Northeast, South, Midwest and West. See below figure 22. This
# Key Factors Elements Descriptions
Average Temperature The average temperature of each state.
Number of Clear Days The number of clear days of each state.
State Population Population in each states.
Racial Population Different racial groups' population in each state.
Home Sales Number of Home Sales in each state.
Household Income Median Household Income in each state.
Environmental Factor
35
would help to understand the different regions and their relationship with sales. These data are affiliated with other three factors, weather, demographic and economic.
Figure 22: Census Regions and Divisions of the United States; Source Census Bureau of United States of America. Source: www.census.com
Weather data are collected through “National Oceanic and Atmospheric Administration or NOAA” including the Average Temperature in 2014 and Number of Clear Days (the average number of days annually when cloud covers at most 30 percent of the sky during daylight hours) from the website of “Current Result”. Both the average temperature of each state and the number of clear days were collected from the same city in the same state.
Demographic data are collected through Wikipedia website which consists of the population of each state in year 2014 and racial breakdown including Non-Hispanic White, Hispanic or Latino, Black, American Indian or Alaskan Native, Asian, Native Hawaiian or Pacific Islander and Mixed Race in fifty different states. The ratios of different races were based on the year in 2010 as it is closest data to 2014.
Economic data are collected from two web sites. The Home Sales (number of home sales in year 2014) in each state is collected from the Zillow web site and the household income by states of year 2014 is collected from the Wikipedia website. The
36
data from Zillow consist from June of 2008 to November of 2015, thus only the data collected from year of 2014 were used.
3.4 Research Analysis
Every data set has a story, and if statistics are used properly, they do a good job of uncovering and reporting that story. In this research, descriptive statistic, correlation coefficient and multiple regressions are used as tools to analyze the data.
Descriptive statistics is taking a set of data and boiling it down to a set of basic information. Correlation coefficient is used as a tool to see the strength and direction of the linear relationship between x and y. Then regression analysis is used to estimate the relationship among variables.
As the researcher collects the data, the researcher arranged data accordingly by each state removing information that is not required such as the name of customer, address, telephone number, etc. Data is then rearranged accordingly into three categories in the format of sales units and dollars. Data are then organized to have all the independent variables including the average temperature, number of clear days, racial population (including Non-Hispanic White, Hispanic or Latino, Black, American Indian or Alaskan Native, Asian, Native Hawaiian or Pacific Islander &
Mixed Race) and the number of Home Sales and Median Household Income with dependent variables mixed into one data.
Lastly, the researcher used both correlation coefficient and multiple regression analysis as tools on the data to provide the results in the next chapter.
37
Chapter 4 Results
This chapter displays the results of this research. The first section of this chapter looks at the independent variables of three factors: environmental, demographic and economic. This section uses descriptive analysis to analyze these key factors. Second section examines the correlation between independent variables and dependent variables. Finally, the last section uses multiple regression analysis as a tool to look at each independent variable and its effects on the dependent variables.
4.1 Descriptive Analysis of Key Factors 4.1.1 Sales Data by State
For the environmental factor, this research studies two independent variables, both the average temperature and the number of clear days. For the geographic, sales data are organized by the sales units and dollars of each state in all three total product categories. The researcher also looked at each three different product categories’ sales units and dollars of each state.
The bar graph in figure 23 is showing both sales units and dollars in total of all three categories in each state. It shows New York has the top ranking in sales units (4479 units) and dollars ($173,806.15), while Pennsylvania is in second place in both sales units (4201 units) and dollars ($173,541.56). Florida has the third place in sales dollars ($166,199.59) but fifth place in sales units (2835 units). Massachusetts is in the fourth place of sales dollars ($152,897.69) but has third place in sales units (3785 units). New Jersey is in fifth place in sales dollars ($130,142.88) but in fourth place in sales units (3072 units).
The bar graph in figure 24 is showing both sales units and dollars in Light Filtering product category of each state. It shows that New York has the top ranking again in both sales units (747 units) and dollars ($16,270.25) while California is in second top position in both sales units (567 units) and dollars ($13,470.19). Illinois is the third ranking in both sales units (549 units) and dollars ($12,120.77). New Jersey is in fourth place in both sales units (513 units) and dollars ($11,121.80).
Pennsylvania is the fifth position in both sales units (486 units) and dollars ($10,603.83).
The bar graph in figure 25 is showing both sales units and dollars in Room Darkening product category of each state. It shows that New York yet again has the top ranking in both sales units (1002 units) and dollars ($25,286.93) while Pennsylvania is in second top position in both sales units (744 units) and dollars ($18,920.41). Illinois is the third ranking in sales units (541 units) but fourth place in
38
sales dollars ($13,834.01). Massachusetts is in fourth place in sales units (521 units) but rank fifth in sales dollars ($13,348.01). California is in fifth place in sales units (513 units) but third place in sale dollars ($14,259.30).
The bar graph in figure 26 is showing both sales units and dollars in Light Blocking product category of each state. It shows that Pennsylvania has the top ranking in sales units (2971 units) but second in sales dollars ($144,617.32).
Massachusetts is in second position in sales units (2816 units) but fourth place in sales dollars ($129,998.73). New York is the third ranking in both sales units (2730 units) and dollars ($132,248.97). Florida is the fourth places in sales units (2266 units) but number one rank in sales dollars ($151,979.38). New Jersey is the fifth position in sales units (2266 units) but Texas is in fifth place in sale dollars ($113,412.40). New Jersey is in sixth place in sales dollars ($105,815.30) and Texas is in eighth place in sales units (1529 units).
The researcher also looked at the average dollars per unit of these states in the top rankings of Light Filtering, Room Darkening, Light Blocking and Total categories.
Their data are organized in the below table 3.
Table 3: Average Dollars per Unit in Light Filtering, Room Darkening, Light Blocking and Total.
l
The data clearly shows these three states, California, Florida and Texas are in the top three average dollars per unit in all four different categories. The data suggest that under the same categories, their average window sizes are larger than other states as larger window covering size is more expensive than the smaller size.
LF Avg.
39
Figure 23: Total of all three product categories and its sales units and dollars of each state of USA.
40 Figure 24: Light Filtering product category and its sales units and dollars of each state of USA.
41
Figure 25: Room Darkening product category and its sales units and dollars of each state of USA.
42 Figure 26: Light Blocking product category and its sales units and dollars of each state of USA
43
4.1.2 Weather Data by State
In the Weather factor, this study looks at two variables; average temperature of each state and their number of clear days in a single year. In figure 27, the average temperature data shows that Florida is in second highest rank. The rest of the four best sales states: New Jersey (27th), Pennsylvania (30th), Massachusetts (32nd) and New York (39th); are nowhere on top ten lists. In the number of clear days, their ranking is as follows: Florida is in 23rd place, Massachusetts is in 28th place, New Jersey is in 33rd place, Pennsylvania is in 40th and New York is in 46th. From the graph, there is not enough to suggest relationship between weather and sales units and dollars.
44 Figure 27: Average Temperature and number of clear days of each state of USA.
45
4.1.3 Demographic Data by State
Demographic Factor is referring to the population in each state. However, the researcher also looked at the racial groups in each state. First, a brief look at the population in each state compared to the total sales units and dollars.
In figure 28, the bar graph shows the population in each state and California has the highest population. In second place is Texas, third place is Florida and New York is in fourth place. Pennsylvania is in sixth place, New Jersey is in eleventh place and Massachusetts is in fourteenth place. The data suggest weak relationship between population and sales units and dollars.
46 Figure 28: Population of each state in USA.
47
4.1.4 Economic Data by State
Economic factor consist of two independent variables which include the number of home sales in a single year in each state and the median household income in a single year in each state.
In figure 29, the bar graph shows that Florida has the top in Home Sales but is in 37th place in Median Household Income. Pennsylvania ranks 6th place in Home Sales and 23rd place in Median Household Income, while New York has 9th place in Home Sales and 16th place in Median Household Income. New Jersey is at 15th position in Home Sales and 10th position in Median Household Income and Massachusetts has 28th place in Home Sales and 6th place in Median Household Income.
48 Figure 29: Number of Home Sales and median Household Income of each state of USA.
49
4.2 Correlation between Studied Variables
In this section is the study of correlation coefficient between independent variables (average temperature, clear days, population in seven different race groups, number of home sales and median household income) and eight dependent variables (sales units and dollars in three categories and total of three categories).
Table 4: Correlation Coefficients between Weather Factors & Sales Variables
Clearly from the data, the environmental variables (both average temperature and number of clear days) do not have a strong relationship with dependent variables, from the table 4, the correlation coefficients are of +0.032, +0.045, -0.023, -0.006, +0.122, +0.227, +0.086 & +0.195 in average temperature and the correlation coefficients are -0.162, -0.151, -0.212, -0.197, -0.174, -0.117, -0.186 and -0.131 in clear days respectively. Note two highest correlation coefficients are Light Blocking in dollars with 0.227 and Room Darkening in units of -0.212. This suggests that the higher the average temperature would have a weak positive linear relationship with sales dollars of Light Blocking category. Also, the higher the number of Clear Days would have a weak negative linear relationship with the sales unit of Room Darkening category.
50
Table 5: Correlation Coefficients between State Populations & Sales Variables
Table 5 shows that there is a strong correlation between state populations and the sales units and dollars in different categories of the product. Table 5 shows correlation coefficients are +0.749, +0.776, +0.663, +0.688, +0.577, +0.655, +0.642 & +0.688 separately. This means that the correlation between population and the dependent variables has a moderate linear relationship and both sales units and dollars of Light Filtering have strong linear relationships. Thus, the result supported that the population has positively correlated to the sales of the product in both units and sales.
Racial groups including Non-Hispanic White, Hispanic or Latino, American Indian or Alaskan Native, Asian, Native Hawaiian or Pacific Islander and Mixed Race are examined with the dependent variables (Light Filtering, Room Darkening, Light Blocking and Total Sales in sales units and dollars).
Population
51 Table 6: Correlation Coefficients between Racial Populations & Sales Variables.
Non-Hispanic White
Hispanic or
Latino Black American Indian or
Alaskan Native Asian Native Hawaiian or
Pacific Islander Mixed race
Light Filtering (Units) 0.824 0.511 0.702 0.187 0.576 0.110 0.590
Light Filtering (Dollars) 0.844 0.545 0.709 0.214 0.605 0.139 0.623
Room Darkening (Units) 0.767 0.404 0.643 0.109 0.478 0.031 0.489
Room Darkening (Dollars) 0.784 0.435 0.661 0.130 0.508 0.057 0.518
Light Blocking (Units) 0.680 0.319 0.698 0.002 0.310 -0.034 0.360
Light Blocking (Dollars) 0.738 0.410 0.796 0.052 0.341 0.010 0.419
Total (Units) 0.745 0.377 0.716 0.050 0.394 -0.001 0.433
Total (Dollars) 0.774 0.436 0.799 0.075 0.390 0.025 0.458
Sales Variables Category 1
Category 2
Category 3
Total Three Categories
52
In table 6, most races have strong and modest linear relationship with three different product categories (Light Filtering, Room Darkening and Light Blocking) and Total Sales in units and dollars. There are two races; American Indian or Alaskan Native and Native Hawaiian or Pacific Islander; showing very weak correlation coefficients, they are +0.187, +0.214, +0.109, +0.130, +0.002, +0.052, +0.050 and
In table 6, most races have strong and modest linear relationship with three different product categories (Light Filtering, Room Darkening and Light Blocking) and Total Sales in units and dollars. There are two races; American Indian or Alaskan Native and Native Hawaiian or Pacific Islander; showing very weak correlation coefficients, they are +0.187, +0.214, +0.109, +0.130, +0.002, +0.052, +0.050 and