Essay

MISSION STATEMENTS AND VISION STATEMENTS:
EXAMINING THE RELATIONSHIP TOWARD

PERFORMANCE OUTCOMES

Jerry Allison, University of Mount Olive

ABSTRACT

This paper examines 798 firms with mission statements and vision statements to show there
are relationships between the two and posits that strong relationships produced greater
organizational performance. Using the taxonomies of Allison (2017a), Allison (2017b), and an
extension of the latter developed in this paper, the statements are classified into their taxonomic
groups and then analyzed statistically. The results surprisingly show a single strong link between
one type of mission statement and one type of vision statement. This paper then discusses how
such a relationship may result in superior performance outcomes. Consequently, this paper
significantly contributes to theory by finding a specific relationship between statements, discussing
why some firms have this relationship, and then extending this discussion to organizational
performance.

INTRODUCTION

Mission statements have been a frequently studied topic (Vizeu & Matitz, 2013). Also
frequently studied has been the topic of vision by virtue of it being a major component of other
subjects such as leadership and strategic management. However, vision statements as a codified
document have not been studied as much as mission statements. Nevertheless, because both
statements are text, rigorous study of them has been arguably difficult. It has been the increase in
computing power that has led to the development of techniques to analyze text such as text
analytics.

Text analytics has provided as way in which to analyze mission statements and vision
statements without researcher bias. Allison (2017a) provided a natural language taxonomy of
vision statements while Allison (2017b) provided a natural language taxonomy of mission
statements. Because the mission and vision statements are from the same organizations, it may be
possible to find relationships between the two and make some conclusions about performance
outcomes. That is why this paper exists.

The purpose of this paper is to explore the relationships between mission and vision
statements from the same organizations and to determine if there are organizational performance
outcomes from those relationships. The uniqueness of this paper is that it utilizes the natural
language taxonomies of textual constructs to study the relationships between those constructs.
Additionally, in order to study these relationships this paper extends the three-class taxonomy of
mission statements provided by Allison (2017b) by dividing the three parent classes into 20 child
classes. Finally, this paper significantly contributes to theory by developing and testing two
hypotheses that show there are relationships between types of mission statements and types of
vision statements and extending these findings to conclusions about performance.

Global Journal of Managment and Marketing Volume 3, Number 1, 2019

1

LITERATURE REVIEW

Mission Statements
One of the first topics found in many strategic management textbooks is that of the mission

statement (e.g. Hill, Jones, & Schilling, 2015; Grant, 2008). Correspondingly, mission statements
have become a commonly studied tool in research (Vizeu & Matitz, 2013). With all the attention
brought to the topic, a common definition of mission statement is not to be found. Two general
perspectives emerge from the literature. The first is that a mission statement is a “container” that
holds several different statements in it. For example, Powers (2012) states that a mission statement
consists of mission, values, vision, and philosophy while Rajasekar (2013) adds internal and
external analysis, strategy implementation, and strategy evaluation. The second perspective is that
a mission statement is a statement of why the firm exists (e.g. Ganu, 2013; David & David, 2008).
This simpler definition is what this paper accepts as the definition. In order to discover a
relationship between mission and vision statements, the first cannot contain the second or the
findings are convoluted. Also, some firms have mission statements but do not have vision
statements. In order to determine a relationship between the two, a company will need to have
both, implying the narrower definition of mission statement is needed.

Research into mission statements has focused upon three areas. One prevalent focus has
been upon what a mission statement should contain. Some research has been performed to
examine included strategic issues such as strategic differentiation (Finley, Rogers, & Galloway,
2001) and strategic purpose (Perfetto, Holland, Davis, & Fedynigh, 2013; Orwig & Finney, 2007).
Other research has been performed to examine content such as firm customers (Peyrefitte & David,
2006), diversity (Barkus & Glassman, 2008), and marketing information (Anitsal, Anitsal, &
Girard, 2012; Anitsal, Anitsal, & Girard, 2013). Other research has been performed to determine
what should be in a mission statement (e.g. Alshameri, Greene & Srivastava, 2012; Pearce &
David, 1987; King, Case & Premo, 2012; King, Case & Premo, 2014). This latter research has
more prescriptive to create a mission statement that creates some form of advantage for the firm.

A second area upon which mission statement research has focused is as a communication
tool. Mission statements have been shown to be a tool to communicate meaning to a receiver (Sufi
& Lyons, 2003; Nous, 2015) but that meaning should be conveyed both to internal and external
stakeholders directly (Amato & Amato, 2002; Biloslavo, 2004; King, Case & Premo, 2013).
Additionally, the communication to internal stakeholders is vital for the mission statement to be
put into practice (Analoui & Karami, 2002; Rajeasekar, 2013). However, some firms take mission
statements further than just a statement of strategic existence. Some firms have used the mission
statement as a means to create organizational impression management to build an image in the
receiver’s mind (David & David, 2003; Peyrefitte, 2012; Khalifa, 2011).

Research has focused on a third area of mission statements by examining the link between
the statement and organizational performance. Several studies have found a relationship between
the statement and organizational performance (e.g. Bart & Hupfer, 2004; Sheaffer, Landau &
Drori, 2008; Alavi & Karami, 2009; Erol & Kanbur, 2014). However, other studies have shown
no relationship between mission statements and performance (e.g. Sufi & Lyons, 2003; Alawneh,
2015). The discrepancy in findings begs the question of what is occurring to produce such results.
One explanation may lie in how, or if, the mission statement is communicated to the organization.
As stated previously, the mission statement must be communicated and adopted by internal
stakeholders in order for the statement to be effective.

Global Journal of Managment and Marketing Volume 3, Number 1, 2019

2

Vision Statements
The literature written about vision statements is as murky as that of mission statements.

Much of the ambiguity rests in how the topic is approached. Research has occurred on vision
because the topic crosses the fields of strategic management, leadership, and organizational
behavior to name a few. In this view, vision has been defined very loosely as organizational
purpose, strategic intent, strategic goals, and a future state of the organization (e.g. Baum, 1994;
Kantabutra & Avery, 2010) while others define vision as simply the future image of the
organization (Brown, 1998; Carver, 2011). However, this organizational vision is ineffective
unless it is communication properly to stakeholders (Baum, 1994; Kantabutra & Avery, 2010;
O’Connell, Hickerson, & Pillutla, 2011) and accepted by stakeholders (Slack, Orife, & Anderson,
2010). One potential way of communicating the vision is through a vision statement.

This paper focuses upon the vision statement as a written communication of organizational
vision. Similar to mission statements, vision statements have also been researched in three areas.
One obvious area is the relationship between a vision statement and organizational performance.
Vision statements have been found to create goals for employees that create better employee
effectiveness and through that customer satisfaction (Kantabutra & Avery, 2010). Organizational
effectiveness can also be enhanced through the use of organizational impression management via
vision statements (Price, 2012). However, all of the positive effects are negated and can be
reversed if the vision statement is just a collection of words and not implemented (Lucas, 1998).

A second area of research has been upon the content of vision statements. Brown (1998)
stated that most vision statements were poorly written or had no focus. Since that time, whether
coincidentally or not, some research has focused upon the elements that need to be within a vision
statement. Some have focused on the length of time (Brown, 1998) while others have focused
upon the statement containing motivating and challenging goals (e.g. Lucas, 1998; MacLeod,
2016; Meade & Rogers, 2001). MacLeod (2016) also adds the caveat that the content of the
statement is irrelevant if the vision is not taken seriously.

The final area of study for vision statements is how these are implemented. Some of this
research has been focused upon ensuring the statement is communicated to internal stakeholders
in order to guide their actions (Lucas, 1998; Payne, Blackbourn, Hamilton, & Cox, 1994).
Kantabutra and Avery (2010) take this notion further and investigate how the statement can be
communicated in order to enhance its effectiveness.

HYPOTHESES

This paper has defined mission statements and vision statements to be separate, individual

statements conveying two different aspects of an organization. This definition was a necessity in
order to focus upon each of those for analysis. However, the literature does not always distinguish
between the two. Some research has tied the two together (e.g. Braun, Wesche, Frey, Weisweiler,
& Peus, 2012; Matejka, Kurke, & Gregory, 1993; Powers, 2012; Rajasekar, 2013; Baum, 1994).
Similarly, strategic management textbooks have considered mission statements to be a “container”
in which mission, vision, values, and other proclamations are kept (e.g. Hill, et al. 2015). Thus,
research does not have clearly defined boundaries for mission and vision. If research cannot keep
the two separate, then practitioners may have an equally difficult time separating the two. This
situation is lamented by Kantabutra and Avery (2010). Thus, when firms are considering creating
these statements, they may be created at the same time, implying a consistent theme during
creation.

Global Journal of Managment and Marketing Volume 3, Number 1, 2019

3

Braun et al (2012) created an extended model of the process of creating the mission
“container” in order to achieve positive organizational results. This model emphasizes the creation
of the statements should be done concurrently. Strategic management texts (e.g. Grant, 2008; Hill,
et al 2015) often place these statements together in the subject presentation, implying these go
hand-in-hand and should be developed at the same time. It is not a far stretch to say that statements
developed at the same time are going to be similar.

Erol and Kanbur (2014) state mission statements and vision statements are pictures of
organizational capability. Fairhurst, Jordan, and Neuwirth (1997) go further and state that these
statements are interdependent so that they project the same organizational characteristics, just in
their own way. Unity of purpose through the use of these statements is one way of meeting
objectives (Gurley, Peters, Collins, & Fifolt, 2015). This unity through these statements can only
be met if the statements themselves are unified. Unity in the statements can only be achieved if
similar or related wording is used in the statements.

Thus, using a taxonomy based on the language in the statements, the conclusion is that an
organization’s mission type and vision type are related. Thus, the following hypothesis can be
made:

Hypothesis 1: There is a significant relationship between the type of mission statement an
organization chooses and the type of vision statement an organization chooses.

Chun and Davies (2001) state companies do not pay enough attention to content in mission

statements. In a similar notion, most vision statements are poorly written (Brown, 1998). Some
organizations do not seem to put the necessary effort into creating these documents to become a
vital component of the organization (Lucas, 1998). Part of the reason for this may be the
organizations do not understand the important nature of these statements. As a natural
consequence, many vision statements are written simply because it is something to do (MacLeod,
2016) and the same conclusion can be made for mission statements (King, et al, 2011).

Because many organizations do not put effort into creating mission and vision statements
but also are encouraged to create them, it would be natural for these organizations to examine
statements from similar firms and either use those as templates. This last statement may be
corroborated in the literature. Firmin and Gilson (2010) found common themes in mission
statements for colleges. Peyrefitte and David (2006) found similarities in mission statements
across industry boundaries. Thus, there may be duplication of types of mission statements and
vision statements due to “copying”.

Because of this “copying”, common organizations would then have the same type of
mission statements and the same type of vision statements. Thus, there should be a relationship
between some mission statement types and some vision statement types. The following hypothesis
can then be made.

Hypothesis 2: There is a relationship between at least one mission statement type and vision
statements type.

METHODOLOGY

One way to test for any relationship between vision statements and mission statements is
to adopt a classification system for both. Both of these constructs have taxonomies developed.

Global Journal of Managment and Marketing Volume 3, Number 1, 2019

4

Allison (2017a) developed a taxonomy for vision statements while Allison (2017b) developed a
taxonomy for mission statements. Each of these is discussed in turn.

Allison (2017a) developed a taxonomy for vision statements where there are two parent
classes and several child classes. The two parent classes are called Spatially Oriented statements
and Achievementcentric statements. The Spatially Oriented class has eight child classes and the
Achievementcentric class has nine child classes.

The vision statement taxonomy of Allison (2017a) has been chosen for several reasons.
First, as of the writing of this paper, it may be the only detailed taxonomy of vision statements in
existence. Second, this taxonomy was created by using the natural language of 798 vision
statements themselves rather than using predetermined classes. This methodology of using natural
language may be a far superior method because the classification relies on characteristics in the
data rather than potential researcher bias (Duarte & Sarkar, 2011; Kuo-Chung & Li-Fang, 2004).
Finally, the methodology of determining the taxonomy was rigorous within the framework just
described. This process started with creating classes from the basic data, using that classification
to determine rules for classification, and then using those rules to classify the data again to
determine a misclassification rate. The parameters of the text analytics software were altered
individually to find the lowest misclassification rate. Misclassification rate was chosen because
the rules developed by the software would be used to classify statements not in the original data.

Allison (2017b) developed a taxonomy for mission statements consisting of three parent
groups: the Producers, the Partners, and the Promoters. This taxonomy was developed using the
mission statements from the same 798 organizations as the vision statements. This taxonomy was
chosen for this study for the same reasons as for the vision statement taxonomy since the same
process was used. The one negative aspect of this taxonomy is that no child classes were developed
for the three parent classes. In order to draw comparisons between the vision statements classes
and the mission statements classes, the mission statement child classes needed to be developed.

Thus, for this paper, the mission statements were divided based upon the parent classes
mentioned previously. For each separate parent class data set, a partition called an unsupervised
classification was derived. From this unsupervised classification rules were derived for the
classification scheme. These rules were then applied to the original parent data set so that a
misclassification rate could be determined. Once the process was created, each parameter was
altered one by one until a minimum misclassification rate was found. The classification of each
of the parent classes was found using this method. The Producers parent class broke up into four
child classes. The Partners parent class divided into twelve child classes. Finally, the Promoters
parent class separated into six child classes. The child classes are described in Appendix 1. The
rules used to classify mission statements are found in Appendices 2 through 5.

Hypothesis 1

The test for Hypothesis 1 used the same sample that formed the mission statement
taxonomy of Allison (2017b) plus the child classes developed here and the vision statement
taxonomy of Allison (2017a). The test was conducted on two different levels. First, the parent
classes for each type of statement were tested to determine if there is a relationship. Second, the
child classes of the parent classes were also tested for a relationship.

For the first test using the parent classes, the Producers, Partners, and Promoters classes of
the mission statement taxonomy were compared to the Spatial Oriented class and the
Achievementcentric classes of the vision statement taxonomy. This data is nominal data, so the
test to be used in this case is the Chi-Square test for independence (Donnelly, 2015; Bluman, 2015).

Global Journal of Managment and Marketing Volume 3, Number 1, 2019

5

This test has been used in other research testing such as business type and lean manufacturing
usage (Nallusamy, 2016), testing consumer organic produce purchases compared with geographic
region (Mrinia & Maharjan, 2015), and testing the relationship of restaurant layout and ambience
(Jana & Chatterjee, 2014).

The Chi-Square test depends upon the cells in the cross tabulation of the variables having
an expected frequency of five or more. Thus, it was imperative to run a frequency table to find
out how many statements are in each of the cells. The frequency table is shown in Table 1.

Since no cell has an expected count of less than five, the Chi-Square test was performed to
determine if there is some form of relationship between the mission statement parent types and the
vision statement parent types. The results are shown in Table 2.

Table 1: Crosstabulation of the parent classes for mission and vision statements
Vision Parent Name Total

Achievement Spatial

Mission Parent
Name

Partners
Count 139 91 230
Expected Count 140.7 89.3 230.0

Producers
Count 225 163 388
Expected Count 237.3 150.7 388.0

Promoters
Count 124 56 180
Expected Count 110.1 69.9 180.0

Total
Count 488 310 798
Expected Count 488.0 310.0 798.0

Table 2: The results of the Chi-Square test of mission and vision

statement parent classes
Value Degrees of

Freedom
Significance

Pearson Chi-Square 6.219a 2 .045
Likelihood Ratio 6.339 2 .042
Number of Valid Cases 798

The Chi-Square test had a p-value of less than 0.05 indicating the test had a significant
result. The results show there is support for Hypothesis 1 on the parent level, that there is a
relationship between the type of mission statement and type of vision statement chosen by an
organization.

The second level of testing was on the child level. Each child type was coded with the
hundreds place denoting the parent class and the tens and ones digits denoting the child class. For
example, a mission statement with a code of 305 meant the third parent class (Promoters) and the
fifth child class under it. A code of 112 meant the first parent class (Partners) and the twelfth child
class under it. Each of these codings were regarded as nominal data. However, when the
crosstabulation was created, there were many cells (343 out of 374) where the expected frequencies
in the cells were less than five. This is shown in Table 3.

Because there are cells that have an expected frequency of less than five, the Chi-Square
test can yield distorted results (Agresti & Finlay, 1997). Fisher’s exact test and its extensions are
designed for tables such as this that have cells of small expected frequencies; however the test was
designed for smaller tables and, when large tables are involved, the use of computational time and
resources can be unrealistic (Agresti & Finlay, 1997; Schlotzhauer, 2009). Because of the size of

Global Journal of Managment and Marketing Volume 3, Number 1, 2019

6

this table, computing resources were not available leaving the Monte Carlo simulation as the last
alternative.

The Monte Carlo simulation is a method for taking a test that can yield results that are
distorted and finding a more accurate result (Tuffery, 2011). The simulation takes a random
sample and runs the test on that sample. This is done many times, with the average of the results
becoming the simulation’s results. Research using the Monte Carlo simulation includes measuring
body absorption of chemicals (Van Landingham, rence, & Shipp, 2004), creating a decision-
making tool for plant capacity expansion (Renna, 2013), and forecasting hotel occupancy
(Zakhary, Atiya, El-shishiny, & Gayar, 2011).

Table 3: Expected frequencies of mission child classes versus vision child classes

With these mission and vision statements, the simulation was run 10,000 times. The final
results are shown in Table 4. The 99% confidence interval for the p-value has an upper bound of
0.0005 indicating significance for this test. The Mantel-Haenszel statistic shows there was not a
significant difference in the results between each of the samples. The results of this test also show
a relationship between mission statement and vision statement child classes and even stronger
support for Hypothesis 1.

Table 4: Monte Carlo results for comparing mission and vision statement child classes
Statistic Degrees of Freedom Value Probability

Chi-square 336 513.0448 <0.0001 Likelihood Ratio 336 448.9864 <0.0001 Mantel-Haenszel 1 2.6737 0.1020 Phi coefficient 0.8018 Contingency coefficient 0.6256 Cramer’s V 0.2005 Global Journal of Managment and Marketing Volume 3, Number 1, 2019 7 Hypothesis 2 As with the previous hypothesis, the goal is to test the hypothesis on two levels. The first level of testing was the parent level where there are three levels of mission statements and two levels of vision statements. Because there are two levels of vision statements, the second hypothesis can be tested using logistic regression with the two levels of vision statements being the dependent variable and the levels of mission statements being the independent variable. Binary logistic regression is applicable when there is a binary dependent variable and at least one independent variable that is continuous, ordinal, or nominal (Tuffery, 2011). This method has been used in other research such as classifying Norway Spruce Saw Logs (Jappinen & Beauregard, 2000) and in studying motorized versus non-motorized transportation in Ireland (son, McMorrow, & Ghosh, 2013). The model results are shown in Table 5. Table 5: Likelihood Ratio Test Results for Global Null Hypothesis (Binary Logistic Regression) Model -2 Log Likelihood Chi-Square DF Sig. Intercept only 1066.223 Final 1060.241 5.9815 1 0.0145 Because the model is significant, the individual levels can then be examined. For this analysis, a step-wise selection approach was used. The results of this are shown in Table 6. Table 6: Level Significance Test Results (Binary Logistic Regression) Parameter Vision Parent Name DF Estimate Standard Error Chi- Square Sig Exp(Est) Intercept Spatial 1 -0.5574 0.0903 40.90 <.0001 0.561 Promoters Spatial 1 0.2176 0.0903 5.81 0.016 1.243 Only the third parent class of mission statements, the Promoters, had a significant result. The table above tells us that the Promoters have a 24.3% greater chance of being identified with a Spatially Oriented vision statement than the others. This is not to say the majority of Promoters are assigned to Spatially Oriented vision statements. Table 1 shows that statement is not true. It only shows there is a greater probability that it will happen. This finding supported Hypothesis 2 on the parent level. Testing Hypothesis 2 on the child level requires using multinomial logistic regression. Multinomial regression is similar to binary logistic regression except it allows for more than two levels of a dependent variable (Tuffery,2011). Thus, the twenty levels of mission statements can be tested against the 17 levels of vision statements. The results of the model test are in Table 7. Table 7: Likelihood Ratio Test Results for Global Null Hypothesis (Multinomial Logistic Regression) Model -2 Log Likelihood Chi-Square DF Sig Intercept only 4369.227 Final 4212.398 156.8295 48 <0.0001 Since the model itself is significant, the individual levels for mission statements can be checked. The coding for mission statements was the same as for Hypothesis 1. Only three of the levels for mission statements were significant – statements 301, 304, and 305. All of these are from the Promoters parent class, the same class that was significant in the parent level test. The results for the individual levels is shown in in Table 8. Global Journal of Managment and Marketing Volume 3, Number 1, 2019 8 Table 8: Level Significance Test Results (Multinomial Logistic Regression) Effect DF Chi- Square Sig Mission301 16 35.3427 0.0036 Mission304 16 39.8244 0.0008 Mission305 16 32.8455 0.0077 To further understand the results, one must look at the individual level results. These results are found in Appendix 6. The table shows the mission 301 type is associated with the vision types 105 and 107 about four times more often than other mission statements except the mission 304 type that is associated with the vision 107 type about 3.4 times more often than other mission statements. The mission 301 type is also associated with the vision statement types 201, 202, 203, 206, and 207 at least 2.2 times more often than other mission statements except the mission statement 304 type that is associated with vision 202 3.6 times more often and vision 207 4.2 times more often. Finally, mission 305 is less often associated with vision 205 than other mission statements by about 61%. Thus, most of the significant relative associations occur between the third parent class of the mission statements (the Promoters statements) and the second vision statement class (the Spatially Oriented statements), supporting the first test for Hypothesis 2. As a result, there seems to be support for Hypothesis 2. ANALYSIS AND DISCUSSION Hypothesis 1 was tested in two different ways. First, a Chi-Square test was run on the three parent levels of mission statements and the two parent levels of vision statements. The test showed significance, providing parent level support for Hypothesis 1. Because of small cell counts in analyzing the child levels, the Chi-Square test could not be performed. However, because of the magnitude of computations, the Fisher Exact test could not be performed either. Consequently, a Monte Carlo simulation was used to take 10000 sample of the table, use the Chi-Square test, and compose a confidence interval for the significance of the test. The confidence interval falls very easily into the level of significance providing support for Hypothesis 1 on the child level. Based on the two tests mentioned, there seems to be good support for the supposition there is a general relationship between mission statements and vision statements. If there were no relationship, then the pairings would occur at about the expected cell frequencies. Thus, there seems to be some form of preferred linking of mission statements types and vision statement types. But the specific relationship cannot be determined using these test. This is where hypothesis two helps. In order to find specific relationships between the mission statement types and the vision statement types, two levels again were tested. On the parent level, since vision statements …