Dr.

P
a
tr
ic
k
M
u
r
p
h
y
School of
Mathematical
 SciencesIntro
duction to
Quantitative
Methods
1Theory and
Concepts
TextbookSeeing
Through
StatisticsbyJessica
UttsDuxbury Press
IMPORTANTMY PART OF
THIS COURSE DOES NOT
USEBLACKBOARD,TOP
CLASS,ONLINE CLASSESOR ANY
VARIATION OF THESE
What do
you
know
about
statistic
s?
I
t
’
s

b
o
r
i
n
g
…
 There are
three kinds of
lies:

•Lies
•Damned Lies
•and
•Statistics
•-Benjamin Disraeli
Simpsons’ episode:

•Homer is questioned about his newly

formed vigilante group

Newscaster:Since your
group started up, petty
crime is down 20%, but
other crimes are up.
Such as heavy sack
beating which is up 800%.
So you’re actually
increasing crime.
Homer:You can make
up statistics to prove
anything.

43% of
people know
that.SurveyR
ate each
person on a
scale of 1 to 10
(1=lowest,
10=highest)on
the following
attributes:•Inte
lligence•Attract
iveness•Nicene
ss•Most Likely
to be a Good
WorkerMisus
e of
Statistics
Statisticsis the science of data.
This involves collecting,
analysing and interpreting
information.
Descriptive Statisticsuses
graphical and numerical
techniques to summarise and
display the information
contained in a dataset.
Inferential Statisticsuses
sample data to make
decisions or predictions
about a larger population of
data Chapter 1The
Beginning
M
o
r
e

D
e
f
i
n
i
t
i
o
n
s
Population:The entire
collection of individuals or
objects about which
information is desired.
Sample:A part (subset) of the
population selected in some
prescribed manner.
Variable:A characteristic or
property of an individual unit in
the population.
Representative Sample:A
selection of data chosen from the
target population which exhibits
characteristics typical of the
population.
Representative samples
should give unbiased
estimates
Mater responds to drug trial controversy 03 October 2005 21:56
Dublin's Mater Hospital has said the requirement of some pharmaceutical companies
that women of childbearing age use contraceptives during trials of potential drug
treatments, goes against its ethos.
The hospital said that individuals and couples have the right to decide themselves
about how they avoid pregnancy.
The Mater was responding to the controversy over the decision of its ethics
committee to defer approval of a clinical trial of a potential lung cancer drug,
because of a requirement that those who participate in the trial use contraception.
Trials of the drug are underway at Tallaght and Beaumont Hospitals.
The hospital said it was committed to meeting all of its legal requirements regarding
clinical trials while at the same time upholding the principles and ethos of the
hospital's mission.
St Vincent's University Hospital cancer specialist Dr John Crown said hospital ethics
committees should not object to the use of contraception, given the potentially fatal
effect of experimental cancer treatment on an unborn child.
About our Live Votes and surveys
How 1,000 people can be more representative than 200,000
One week in the middle of the Clinton-Lewinsky scandal, more than
200,000 people took part in an MSNBC Live Vote that asked whether
President Clinton should leave office. Seventy-three percent said yes.
That same week, an NBC News-Wall Street Journal poll found that only 34
percent of about 2,000 people who were surveyed thought so. To explain
the vast gap in the numbers in this and other similar cases, it is
necessary to look at the difference in the two kinds of surveys.
POLLS Journalists use polls to gauge what the public is thinking. The
most statistically accurate picture is captured by using a randomly
selected sample of individuals within the group that is being targeted,
typically adult Americans.
While a poll of 100 people will be more accurate than a poll of 10, studies
have shown that accuracy begins to improve less at about 500 people and
increases only a minor amount beyond 1,000 people.
So, in the case of that NBC-WSJ poll, only 2,005 adults were surveyed by
the polling organizations of Peter D. Hart and Robert M. Teeter. The poll
was conducted by telephone and had a margin of error of plus or minus
2.2 percentage points at the 95 percent confidence level. The confidence
level means that if the same poll were conducted 100 times, each one
randomly selecting the people polled, only five of the polls would be
expected to yield results outside the margin of error.
Random selection of those polled is necessary to ensure a broad
representation of the population at large. For example, a nationwide poll
asking which NBA team is the best would likely yield a far different
answer in Philadelphia than in Los Angeles. (And neither one would be a
good sample of the population at large.)
In the NBC-WSJ survey, pollsters first randomly selected a number of
geographic areas and then telephone numbers were generated in a way
that allowed all numbers in those areas (both listed and unlisted) an
equal chance to be called. Only one adult in each household was then
selected to answer the poll.
While variation can occur depending on what questions are asked and
how they are asked, similar questions tend to yield similar answers. One
way to account for variation, however, is to ask the same question over a
period of time.
ONLINE SURVEYS In contrast, MSNBC's online surveys (Live Votes) may
reflect the views of far more individuals, but they are not necessarily
representative of the general population.
To begin with, the people who respond choose to do so — they are not
randomly selected and asked to participate, but instead make the choice
to read a story about a certain topic and then vote on a related question.
There is thus no guarantee that the votes would reflect anything close to
a statistical sample, even of MSNBC.com users: The participants in a
Sports Live Vote and a Politics Live Vote may overlap, but each group is
likely to be dominated by people with an interest in each particular area.
In addition, while MSNBC.com’s Live Votes are designed to allow only one
vote per user, someone who wants to vote more than once could simply
use another computer or another Internet account.
According to Nielsen//NetRatings, nearly 75 percent or 204.3 million
Americans had access to the Internet from home in early 2004. In
contrast, more than 90 percent of Americans live in homes with a
telephone.
This does not mean that Internet polling cannot be scientific. Harris
Interactive, for example, has set up a system with checks and balances
that allow it to use the Internet to obtain survey results comparable with
more traditional methods.
But MSNBC’s Live Votes are not intended to be a scientific sample of
national opinion. Instead, they are part of the same interactive dialogue
that takes place in our online chat sessions: a way to share your views on
the news with MSNBC writers and editors and with your fellow users. Let
us know what you think.
 Searching for Correlations Between Beauty and
Intelligence From the Male Perspective: A Brief Study
Alanna Albano
Abstract: In order to determine if men place particular stereotypes on women based on looks
alone, five members of the Bryn Mawr College community answered a brief survey. The
survey involved looking at a collection of pictures of different women, and then answering a
series of questions that made references to the women's beauty, intelligence, personality,
interactions with others, and job capability. Some small specific trends were observed in the
completed surveys; however, it was rather disappointing to find that no two surveys were
exactly alike in their responses about the women. Although no definitive conclusion was
reached in regards to associating certain stereotypical roles with a particular feminine
appearance, the majority of the survey responses reinforced some of the beauty-related issues
previously debated in class.
Keywords: Beauty, symmetry, confidence, intelligence
Experimental Procedure: Nine color pictures of women from various age, job, and
socioeconomic divisions were collected onto a single sheet. No information about the women
was provided to the participants other than the pictures. A survey consisting of nine questions
was designed to ask about the intelligence, personality, and status of the women (see
Supporting Information for a listing of all survey questions). Participants were chosen based
on accessibility, relation to the surveyor, and willingness to complete the survey. Candy and
chocolate was sometimes used as an incentive, along with a pleasant disposition and firm,
polite manner in order to attract survey participants. Usually, stating the overall purpose of
the survey worked best when encouraging others to complete the survey. At this particular
time of the year, it was very difficult to find survey participants. A couple of men "politely"
refused the request to complete the survey due to excessive amounts of work and other
commitments. Five members of the Bryn Mawr College community completed the survey.
All participants were men; four were post baccalaureate students between the ages of 24-30,
and one was a housekeeper of the College. Participants required 15-25 minutes to complete
the survey.
Results: The results of the survey were found to be highly varied among the participants, with
only a few common responses to some of the questions. This was probably due to the highly
subjective nature of the survey; most of the participants expressed different reasons for
answering the survey questions in the way that they did. Pictures 3 and 5 got the most votes
for being beautiful due to reasons that included "nice eyes, lips, and hair," "3 looks good with
a scarf on," "a put-together look and incredible/nice eyes," "looks good from many angles,
well-proportioned, nice symmetry in her face," and a look of cleanliness. Picture 7 also
received a majority of votes for beauty for the same reasons
regarding eyes, facial symmetry, and a clean look; however, one participant remarked that
she did not look as put-together as 3 and 5, but still had a natural beauty to her. The general
consensus among the participants was that the woman in picture 9 was the least beautiful.
Her irregular features were found to be quite distracting, and her facial expression was
considered "interesting." 4 and 6 also got a considerable number of votes for not being
beautiful for reasons similar to those for 9. One man thought all of the women were beautiful
except for 4 and 9, and suggested a makeover would help 4. Another commented that 1 and 2
could be beautiful with the help of a different style. One man only chose 3, 5, and 7 to be
beautiful, and said that the rest of the women were not beautiful due to odd features or
looks/styles. Another commented that the reason he thought certain women to not be
beautiful was due to today's societal attitudes toward beauty. He continued to write that 1 and
4 looked like they came from the 1980s, and since the 80s style is outdated, those particular
women did not look beautiful to him.
Most of the participants agreed that 2 and 7 were the most intelligent. It was commented that
2 looked intelligent because of her glasses, shorter hair style, smile, personable expression,
and overall look of a business woman. Another remarked that 7 appeared intelligent due to
her confidence and composure, which gave the impression of a very productive person.
Pictures 6 and 9 were thought to be the least intelligent. One man considered 6 to be
unintelligent because she did not look like she was in control of her life. 9 was deemed as
unintelligent because one participant remarked that she portrayed the stereotypical flaky
blonde. One young man made a very interesting comment regarding the intelligence of 3 and
5. He thought that they were unintelligent because they looked like women from the 1950s,
and he felt that women in that time period were not as focused on higher education. He
admitted that he was probably wrong and that social aspects might be affecting his view; on
the other hand, he did remark that he was giving his honest opinion about pictures 3 and 5.
2, 7, and 8 were voted the most likely to be hired for a job. One man stated he would hire 2
and 8 as fashion consultants, and that it would not be possible to hire 7 because she would
already be the boss. 7 also looks sophisticated enough to be the boss. 2 would also be hired
because she looks successful, seems comfortable around money, is in control of who she is,
looks put-together, and appears intelligent. These same reasons also applied to hiring 7;
another man remarked that he would hire 7 specifically because she is very beautiful and
appears nice, as well as intelligent. The friendly looking appearance of 8 would greatly
increase her chances of being hired, according to one young man. Some interesting
comments were made as to why the other women would not be hired. 1 looks like she is
stuck in the 80s and is against change, which is an important quality for hiring. 4 would not
be hired because she looks socially awkward. 6 is unfit for employment due to her emotional
state, and because she does not look intelligent. Since she does not present herself as
intelligent, 9 would not be hired either. One man stated that who he would hire would
entirely depend on the job that the woman was applying for.
According to one participant, he determined the niceness of the women by their facial
expressions, and another made judgements based on the pictures of women who resembled
nice women in his own life. 2, 4, 7, and 8 got the most votes for being nice
people. Reasons for these choices included soft and welcoming facial expressions, as well as
looking family oriented.1, 3, and 9 were voted to be not nice. 1 and 9 appear to be on the
defensive and look unhappy. 3 would not be nice because people that are pretty often tend to
have an attitude toward others, and expect everything to be given to them. 5 and 6 got mixed
reviews in terms of niceness. One man felt that 5 would be nice only to do something sinister
later, and another thought 5 was nice in the same way that 2 would be. 6 could be nice
because she appears family oriented and emotional, implying that she is a very caring person;
on the other hand, she might be mean since she looks unhappy in the picture.
Most of the men would be likely to give money to 6 if she asked them for it, because she has
a child, looks like she is sad and poor, and seems to really need the money. However, one
man remarked that he would not give 6 any money because she looks all disheveled, and
probably would not put the money to good use. With the exception of 1, 6, and 9, this same
man also stated that he would be more likely to give the other women money because they
look like they try hard in life. One participant would give money to 4 because she looks like
she needs it, and another said that he would actually give money to 9 because she has a cup
(in the picture) to put it in. In terms of lending the women grocery money, two of the men
would lend money to any of the women; one specifically stated he would lend to all of them
because he knew he would feel sorry for them, and at least he would know what the money
was being used for. Another would give money to some of the women, and another said that
he would give it to all of the women except 9, because she does not look like someone who
helps others. One participant made it clear that he would give all of the women money at the
grocery store because money is not important to him.
The women most likely to be offered assistance if they found themselves lost at a busy street
corner were 1, 2, 4, 7, and 8. 9 looks too mean and scary, 6 looks like she is crazy and
unstable, and 3 and 5 look like actors who would only "act" like they were lost at a street
corner. Some of the other participants stated that they would assist any of the women. One
man wrote that he likes to help people in need, especially when they do not ask for it.
Another would help any of them because he would want directions if he were lost, and
someone else said he would give any of them directions if they looked disturbed enough.
All of the participants gave various responses for the types of jobs and lifestyles that they
could envision each of these women living. The summary is as follows: 1 - Ex-rock star from
the 80s, professor, waitress, lawyer, chef, dental hygienist; 2 - Doctor, TV anchor, professor,
mother, lawyer, waitress, comedian; 3 - actress, model, entertainer, professor; 4 - farmer's
wife, professor, lawyer, librarian, scientist, mother, housekeeper; 5 - actress, professor,
mother, housewife; 6 - mother, poor immigrant, housekeeper, professor, housewife,
counselor, social worker; 7 - maintenance worker, housekeeper, writer, CEO, professor,
mother, real estate agent, lawyer; 8 - housewife, mother, professor, social worker, office
secretary, lawyer, housekeeper; 9 - entertainer, waitress, "gold digger"/seeker of men with
money. One man commented that most of the women seemed
capable of a variety of different jobs, although wearing glasses made for better doctors and
scientists.
Discussion: In my last paper, I had written about a beauty survey on feminine beauty and
intelligence that I conducted on twelve female participants from the Bryn Mawr community.
This detailed survey was done in response to a similar, much shorter type of survey that had
been given in class. When the question of whether a beautiful woman pictured on the slide
projector screen was also intelligent, many of the students responded no. This answer
disturbed me greatly, and I decided to find out how different members of the Bryn Mawr
community viewed women's beauty and its connections to intelligence. My first survey did
not contain any male input. Therefore, this motivated me to conduct a second survey in
which the participants were male students and employees of the Bryn Mawr community. The
exact same survey questions and pictures were used for the male participants as for the
previous female ones. Although the small number of five participants did not accurately
represent all male perspectives on feminine beauty and intelligence, it was still very helpful
and interesting to read what they had to say on the topic. Despite some of the many flaws in
the structure and design of the survey, some trends were indeed observed in the survey
responses.
Symmetry, an air of confidence, and a neat, clean appearance played huge roles in
determining which women were most beautiful (3, 5, 7). Asymmetrical features, an unhappy
countenance, and outdated styles were characteristics shared by the least beautiful women, 4,
6, and 9. Of the women declared most beautiful, only 7 would be considered intelligent and
hired for a job. The confidence and beauty of 7 give her an edge in these areas, whereas the
glasses and hairstyle that 2 wears are what label her as intelligent and job worthy. 8 seems
more likely to be hired because of her glasses and friendliness. Somehow, wearing glasses
gives a person the look of an academician. 6 and 9, considered to be the least beautiful, were
also considered to be the least intelligent and not very likely to be hired for a job. Although 3
and 5 were considered beautiful, most of the participants knew that these women were
actresses in real life, which probably influenced their decisions to not specifically label them
as intelligent. It is interesting to consider how responses toward 2 and 7 would have changed
had the participants known that one woman is a model and the other is an actress.
Interestingly enough, the women considered most intelligent and job competent were also
considered to be the nicest people. Ironically enough, the women on the two extremes of the
beauty scale (3, 5, and 9) were deemed to be mean people. 6, who was considered to be one
of the less beautiful women, would be the most likely to receive money if she asked for it
because she has a child and appears sad and distraught. Some opinions about giving her
money might have changed had her child not been in the picture with her. 9 is the least likely
to receive any money because she looks mean and scary. Most of the women, regardless of
their looks, would still receive help from the male participants if they got lost.
I thought it very interesting that only the male housekeeper survey participant thought that
one of the women might be a doctor; none of the male postbacs labeled any of the
women as potential doctors. One would think that the postbacs, who are studying to attend
medical school someday, would be the more likely ones to label some of the women pictured
in the survey as physicians. I noticed that the term "housewife" was mentioned frequently as
a potential career for some of the women, especially 4, 5, 6, and 8. The job associations that
the men made for each of the women were too widespread to make any definitive
connections between being beautiful and having a particular career. However, it should be
noted that all women except 9 were labeled as potential professors, all except 1 and 9 were
considered to be potential mothers, and 1, 2, 4, 7, and 8 were thought to be possible lawyers.
One male participant commented that he judged the jobs and lifestyles of these women based
on the women he knew in his own life; for example, he thought 8 could be an office
secretary, since she looked very much like the secretaries who work in his father's office. He
also thought 1 could be a dental hygienist, since she looks so much like his own. Sometimes
the articles of clothing or jewelry that the women were wearing influenced the men's decision
regarding what kind of lifestyle each women might be living. 2 was thought to live a nice
lifestyle because of her nice glasses. 4 was thought to be a lower middle class person because
of her outdated clothes and glasses. The flashy necklace that 9 wears implied a superficial
lifestyle, as remarked by one male participant.
This survey yielded some very eye-opening data. However, a similar survey conducted on a
much larger scale, and perhaps with a different type of answer response format, would be
more useful in understanding how men view feminine beauty and intelligence. One male
participant suggested giving the survey to more people both on the campus and off the
campus. I heartily agree that this would be a better step in the direction of learning more
about the masculine perspective of beauty. However, if such a survey on a larger scale were
ever conducted, I would hope that the majority of men would be much more willing to
complete it.
Supporting Information: The following are the original questions asked in the survey: Which
of these women do you find to be beautiful? Not beautiful? Either way, explain why. Which
of these women do you find intelligent? Not intelligent? Either way, explain why. Which of
these women would you hire for a job? Why? Which of these women do you think would be
nice people? Which do you think would be mean people? Explain why. If all of these women
asked you for money, which women would you choose to give it to? Why? What kinds of
jobs and lifestyles do you envision each of these women living? Why? If any of these women
found themselves a couple dollars short at the grocery counter, would you spare them some
cash? If so,which women, and why? If you saw any of these women looking hopelessly lost
on a busy street corner, would you immediately offer assistance? Which women and why?
Which of these women do you think would make good Doctors? Lawyers? Mothers? Chefs?
CEOs? Scientists? Models? Actresses? Professors/teachers? Waitresses?

Definition:
Collection of mathematical and statistical methods used in the solution of managerial and
decision-making problems, also called operations research (OR) and management
science. There are numerous tools available under these headings such as Linear
Programming (LP), Economic Order Quantity (Eoq), Learning Curve theory, Pert and
Regression Analysis.
Advantages And Disadvantages Of The
Qualitative Research Methods
Qualitative research is used to denote approaches which are supported by a set of
hypotheses concerning the way the social world functions. It deduces many of its basic
principles from the perspective that there are fundamental differences between the
science of human world and science of natural world and consequently needs to use
distinctive methods. Here, attention is focused upon looking at the world through the eyes
of studied objects and upon evolving concepts and theories which are grounded in the
collecting data. So, qualitative research connected with own accounts of the individuals
of their attitudes and behavior.

The significance of qualitative research consists in setting stress on describing,

understanding complex phenomena. It investigates, for instance, the relationships and
patterns among factors or the context in which the activity happens. It is concentrated on
understanding the full many-dimensional picture of the subject of investigation.

The approaches of the qualitative research differ from the methods of the quantitative
research. Quantitative methods have their aim in dividing into clearly defined parts, or
variables. When we research an issue which we know how to quantify, for example, what
can be quantified for sure, we may leave out the factors which are crucial to the real
understanding of the phenomena under study. Disadvantage of the quantitative as well as
qualitative research is that they do not always underpin understanding of multi-
dimensional pictures.

Qualitative methods are helpful not only in giving rich explanations of complex
phenomena, but in creating or evolving theories or conceptual bases, and in proposing
hypotheses to clarify the phenomena. Besides, value of the qualitative research consists in
validity of the information received; people are minutely interviewed so as the obtained
data would be taken as correct and believable reports of their opinions and experiences.
Its major disadvantage is that small group of interviewed individuals can not be taken as
representative.

THE ADVANTAGES OF EMPLOYING QUANTITATIVE AND QUALITATIVE

METHODS IN INTERCULTURAL RESEARCH: PRACTICAL IMPLICATIONS
FROM THE STUDY OF THE PERCEPTIONS OF INTERCULTURAL
COMMUNICATION COMPETENCE BY AMERICAN AND RUSSIAN
MANAGERS

Alexei V. Matveev
(New York, USA)
Published: Collected research articles, Bulletin of Russian Communication Association
"THEORY OF COMMUNICATION AND APPLIED COMMUNICATION", Issue 1 /
Edited by I.N. Rozina, Rostov-on-Don: Institute of Management, Business and Law
Publishing, 2002. - 168 p. P. 59-67

Abstract
Intercultural researchers usually use two types of investigation processes. First is
quantitative research, which employs numerical indicators to ascertain the relative size
of a particular communication phenomenon. The second type of investigation process is
qualitative research, which employs symbols and words to indicate the presence or
absence of phenomena or categorize them into different types. Quantitative and
qualitative observations provide intercultural researchers with different ways of
operationalizing and measuring theoretical constructs and practical concepts. While
quantitative methods can provide a high level of measurement precision and statistical
power, qualitative methods can supply a greater depth of information about the nature of
communication processes in a particular research setting.

This research investigated the perception of intercultural communication competence by

American and Russian managers with experience on multicultural teams. The researcher
employed a survey method to solicit information about intercultural communication
competence and multicultural team performance from 124 American and Russian
managers. Qualitative interviews of 40 American and Russian managers enriched the data
collected by using questionnaires. This study yielded a number of conclusions about the
importance of intercultural communication competence in the performance of
multicultural teams.

The researcher experienced a number of advantages of applying both quantitative and

qualitative methods in this intercultural research. Quantitative methods ensured high
levels of reliability of gathered data. Qualitative research allowed for obtaining more in-
depth information about how the managers perceive intercultural communication
competence and its relationship with the performance of multicultural teams. This article
is useful for empirical researchers, qualitative scholars, and business practitioners who
are engaged in intercultural research or who plan to perform a future study.

The Research Problem

In today's world, people spend a considerable time communicating across cultures with
other people and organizations around the world. Intercultural communication
competence of employees of multicultural organizations has risen to a higher level of
importance. Today a large portion of communication in multicultural organizations
occurs between people with different cultural backgrounds (Hofner Saphiere, 1996).
Team members with different cultural backgrounds are likely to have difficulties in
understanding each other because of the difference in the perception about what
competent communication is. Past research shows that competent communication and an
effective team development process can address many of the challenges of multicultural
teams (Christophel, 1996; Gudykunst & Kim, 1984; Hofstede, 1991; Lewis, 1998;
McCroskey & McCroskey, 1986; McCroskey & Richmond, 1990; Parker, 1998;
Rhinesmith, 1983, 1996; Young, 1998).

To work effectively with diverse people, managers of multicultural organizations need to

know not only about the culture of the person with whom they are interacting, but also
about his or her personality, behavior patterns in conflict situations, demographics, and
life experiences. An ability to work in a diverse workforce requires managers to become
competent intercultural communicators. The research presented in this paper analyzed the
perception of intercultural communication competence by American and Russian
managers with experience on multicultural teams. The primary goals of this study were to
(a) investigate how American and Russian managers perceive intercultural
communication competence, and (b) describe a relationship between intercultural
communication competence of managers and the performance of multicultural teams.

Comparison of Quantitative and Qualitative Methods of Inquiry

Intercultural researchers are deemed to have certain challenges due to cultural, linguistic,
business practice, and communication differences of the research participants, survey
respondents, and interviewees. Combing quantitative and qualitative methods helps to
prevent some of these challenges and manage others. The following section highlights
strengths and weaknesses of the quantitative and qualitative methods experienced by
intercultural researchers.

Quantitative Method

The functional or positivist paradigm that guides the quantitative mode of inquiry is
based on the assumption that social reality has an objective ontological structure and that
individuals are responding agents to this objective environment (Morgan & Smircich,
1980). Quantitative research involves counting and measuring of events and performing
the statistical analysis of a body of numerical data (Smith, 1988). The assumption behind
the positivist paradigm is that there is an objective truth existing in the world that can be
measured and explained scientifically. The main concerns of the quantitative paradigm
are that measurement is reliable, valid, and generalizable in its clear prediction of cause
and effect (Cassell & Symon, 1994).

Being deductive and particularistic, quantitative research is based upon formulating the
research hypotheses and verifying them empirically on a specific set of data (Frankfort-
Nachmias & Nachmias, 1992). Scientific hypotheses are value-free; the researcher's own
values, biases, and subjective preferences have no place in the quantitative approach.
Researchers can view the communication process as concrete and tangible and can
analyze it without contacting actual people involved in communication (Ting-Toomey,
1984).

The strengths of the quantitative method include:

 • Stating the research problem in very specific and set terms (Frankfort-Nachmias
& Nachmias, 1992);
• Clearly and precisely specifying both the independent and the dependent variables
under investigation;
• Following firmly the original set of research goals, arriving at more objective
conclusions, testing hypothesis, determining the issues of causality;
• Achieving high levels of reliability of gathered data due to controlled
observations, laboratory experiments, mass surveys, or other form of research
manipulations (Balsley, 1970);
• Eliminating or minimizing subjectivity of judgment (Kealey & Protheroe, 1996);
• Allowing for longitudinal measures of subsequent performance of research
subjects.

The weaknesses of the quantitative method include:

• Failure to provide the researcher with information on the context of the situation
where the studied phenomenon occurs;
• Inability to control the environment where the respondents provide the answers to
the questions in the survey;
• Limited outcomes to only those outlined in the original research proposal due to
closed type questions and the structured format;
• Not encouraging the evolving and continuous investigation of a research
phenomenon.

Qualitative Method

Qualitative research shares the theoretical assumptions of the interpretative paradigm,

which is based on the notion that social reality is created and sustained through the
subjective experience of people involved in communication (Morgan, 1980). Qualitative
researchers are concerned in their research with attempting to accurately describe,
decode, and interpret the meanings of phenomena occurring in their normal social
contexts (Fryer, 1991). The researchers operating within the framework of the
interpretative paradigm are focused on investigating the complexity, authenticity,
contextualization, shared subjectivity of the researcher and the researched, and
minimization of illusion (Fryer, 1991).

Qualitative research in general is more likely to take place in a natural setting (Denzin,
1971; Lincoln & Guba, 1985; Marshall & Rossman, 1989). This means that topics for
study focus on everyday activity as "defined, enacted, smoothed, and made problematic
by persons going about their normal routines" (Van Maanen, 1983, p. 255). Qualitative
research is less likely to impose restrictive a priori classification on the collection of data.
It is less driven by very specific hypotheses and categorical frameworks and more
concerned with emergent themes and idiographic descriptions (Cassell & Symon, 1994).

Extending the fundamental beliefs of the interpretative paradigm, one can name three
characteristics of qualitative inquiry. First, qualitative research is the study of symbolic
discourse that consists of the study of texts and conversations. Second, qualitative
research is the study of the interpretive principles that people use to make sense of their
symbolic activities. Third, qualitative research is the study of contextual principles, such
as the roles of the participants, the physical setting, and a set of situational events, that
guide the interpretation of discourse (Ting-Toomey, 1984).

The strengths of the qualitative method include:

• Obtaining a more realistic feel of the world that cannot be experienced in the
numerical data and statistical analysis used in quantitative research;
• Flexible ways to perform data collection, subsequent analysis, and interpretation
of collected information;
• Provide a holistic view of the phenomena under investigation (Bogdan & Taylor,
1975; Patton, 1980);
• Ability to interact with the research subjects in their own language and on their
own terms (Kirk & Miller, 1986);
• Descriptive capability based on primary and unstructured data;

The weaknesses of the qualitative method include:

• Departing from the original objectives of the research in response to the changing
nature of the context (Cassell & Symon, 1994);
• Arriving to different conclusions based on the same information depending on the
personal characteristics of the researcher;
• Inability to investigate causality between different research phenomena;
• Difficulty in explaining the difference in the quality and quantity of information
obtained from different respondents and arriving at different, non-consistent
conclusions;
• Requiring a high level of experience from the researcher to obtain the targeted
information from the respondent;
• Lacking consistency and reliability because the researcher can employ different
probing techniques and the respondent can choose to tell some particular stories
and ignore others.

The Research Methods Used in This Study

Investigating how different cultures perceive intercultural communication competence

and its relationship with multicultural team performance included field research in Russia
and data collection in the United States. The participants in this study were 124 upper and
middle managers of medium and large international companies. The researcher selected
the companies and the actual research participants based on the scale of their
international operations, their history of working in intercultural teams, and their
willingness to participate in this research study.

The researcher employed the Integrated Intercultural Communication Competence Model

(Model) to examine the perception of intercultural communication competence by
American and Russian managers with experience on multicultural teams. The Model is
based on the research of Abe and Wiseman (1983) and Cui and Awa (1992). The
theoretical foundations of the Model are rooted in the cultural studies of Edward Hall
(1956, 1976), Hofstede's (1980, 1991) cultural dimensions of power distance,
individualism-collectivism, uncertainty avoidance, and masculinity-femininity, and the
communication orientations across cultures (Gudykunst & Ting-Toomey, 1988). The
Model views intercultural communication competence of a team member as being based
on four underlying dimensions: interpersonal skills, team effectiveness, cultural
uncertainty, and cultural empathy. The factor analysis of the pilot data confirmed these
four dimensions as being critical characteristics for members of multicultural teams
(Matveev, Rao, & Milter, 2001).

The researcher used both quantitative and qualitative methods to solicit information from
the respondents on intercultural communication competence and multicultural team
performance. The quantitative part of the research included two research questionnaires.
The Intercultural Communication Competence Questionnaire with the internal
consistency reliability of Chonbach alpha of 0.88 was based on the Integrated
Intercultural Communication Competence Model (Matveev, Rao, & Milter 2001). The
High Performance Team Questionnaire with the test-retest reliability of .82 and the
internal consistency alpha of .88 was based on the integrated model of group
development (Wheelan, 1990, 1994, 1999).

The qualitative part of the research included 40 semi-structured interviews with the
American and Russian managers. The goal of the interviews was to obtain more in-depth
information about how the managers perceive intercultural communication competence
and its relationship with the performance of multicultural teams (Frankfort-Nachmias &
Nachmias, 1992). The researcher asked the American and Russian managers to describe
an interculturally and communicatively competent employee, to rank the dimensions of
intercultural communication competence based on their understanding of this theoretical
construct, and to describe their views about a relationship between intercultural
communication competence and performance in multicultural teams.

ntegrating Quantitative and Qualitative Methods in Social Marketing Research

by Nedra Kline Weinreich

Introduction

Traditionally, research in the field of health promotion has followed in the footsteps of its
"older brother," medicine. However, the reductionistic model of disease causation cannot
adequately describe the complex mechanisms that influence health behavior. Social
marketers working to promote health have learned that rigorous quantitative research
surveys do not necessarily provide all of the data needed to develop effective
communications. Consequently, qualitative methods such as focus groups and in-depth
interviews, as well as less precise but useful semi-quantitative approaches, such as
intercept surveys, have emerged as part of their research repertoire. In an ideal social
marketing program, researchers use both quantitative and qualitative data to provide a
more complete picture of the issue being addressed, the target audience and the
effectiveness of the program itself. The purpose of this paper is to look at how these two
different research approaches can be integrated to inform the development of an effective
social marketing program.

Qualitative and Quantitative Methods: A Comparison

An examination of the quantitative and qualitative paradigms will help to identify their
strengths and weaknesses and how their divergent approaches can complement each
other. In most cases, researchers fall into one of the two camps--either relying exclusively
upon "objective" survey questionnaires and statistical analyses and eschewing warm and
fuzzy qualitative methods, or using only qualitative methodologies, rejecting the
quantitative approach as decontextualizing human behavior. However, social marketing
researchers recognize that each approach has positive attributes, and that combining
different methods can result in gaining the best of both research worlds.

Quantitative research uses methods adopted from the physical sciences that are designed
to ensure objectivity, generalizability and reliability. These techniques cover the ways
research participants are selected randomly from the study population in an unbiased
manner, the standardized questionnaire or intervention they receive and the statistical
methods used to test predetermined hypotheses regarding the relationships between
specific variables. The researcher is considered external to the actual research, and results
are expected to be replicable no matter who conducts the research.

The strengths of the quantitative paradigm are that its methods produce quantifiable,
reliable data that are usually generalizable to some larger population. Quantitative
measures are often most appropriate for conducting needs assessments or for evaluations
comparing outcomes with baseline data. This paradigm breaks down when the
phenomenon under study is difficult to measure or quantify. The greatest weakness of the
quantitative approach is that it decontextualizes human behavior in a way that removes
the event from its real world setting and ignores the effects of variables that have not
been included in the model.

Qualitative research methodologies are designed to provide the researcher with the
perspective of target audience members through immersion in a culture or situation and
direct interaction with the people under study. Qualitative methods used in social
marketing include observations, in-depth interviews and focus groups. These methods are
designed to help researchers understand the meanings people assign to social phenomena
and to elucidate the mental processes underlying behaviors. Hypotheses are generated
during data collection and analysis, and measurement tends to be subjective. In the
qualitative paradigm, the researcher becomes the instrument of data collection, and
results may vary greatly depending upon who conducts the research.

The advantage of using qualitative methods is that they generate rich, detailed data that
leave the participants' perspectives intact and provide a context for health behavior. The
focus upon processes and "reasons why" differs from that of quantitative research, which
addresses correlations between variables. A disadvantage is that data collection and
analysis may be labor intensive and time-consuming. In addition, these methods are not
yet totally accepted by the mainstream public health community and qualitative
researchers may find their results challenged as invalid by those outside the field of social
marketing.

Social Marketing Research

The traditional health promotion professional conducts research at the beginning of a

project to develop an intervention, and again at the end to evaluate the effectiveness of
the intervention. In contrast, social marketers utilize research throughout the planning,
development, implementation and evaluation phases of the program; social marketing is a
process of continuous development and testing. Many of the tools used to develop social
marketing programs--focus groups, consumer marketing databases, intercept surveys--
have their origins in the field of commercial market research, and are based on "what
works" for gathering various types of needed data. Social marketing relies upon
consumer-focused research to learn as much about the target audience as possible by
looking at their lives from many different angles--both quantitatively as part of a larger
group and qualitatively to investigate individual attitudes, reactions, behaviors and
preferences.

Social marketing programs use research throughout the life of a project. Research in
social marketing is conducted specifically to help make better decisions at key points in
the process (Andreasen, 1995). These decisions may include which target audience,
messages and media to choose; whether to make changes in program strategy during
implementation; and whether to continue the program. Pinpointing the facts needed to
make these decisions will help to identify the best methods for subsequently collecting
this data. Some types of information may require quantitative data collection methods,
such as detecting any measurable differences in knowledge or behaviors once the
program has been implemented. Soliciting audience reactions to a selection of program
messages, on the other hand, may be best done through qualitative methods. An effective
and responsive program requires a combination of research approaches in order to have
the data needed for decision making.

Professionals who come to social marketing from a traditional health promotion

background may have a difficult time in reconciling their notion of "what research is"
with some of the methods that social marketers have appropriated from the commercial
marketing tool kit. Even those who are committed to using a mix of research methods
may encounter institutional resistance to deviating from the quantitative paradigm,
particularly when the proposed research will occur in a governmental or academic
setting. However, as the field of health promotion evolves from a focus on individual
lifestyles and risk factors to a broader concept of social and environmental factors
influencing morbidity and mortality, researchers must employ a variety of methods to
reflect this new perspective.
Toward an Integrative Social Marketing Research Model

As a useful starting point, Steckler et al. (1992) have delineated four possible models of
integrating qualitative and quantitative methods in health education research. In the first
approach, qualitative methods contribute to the development of quantitative instruments,
such as the use of focus groups in questionnaire construction. The second model consists
of a primarily quantitative study that uses qualitative results to help interpret or explain
the quantitative findings. In the third approach, quantitative results help interpret
predominantly qualitative findings, as when focus group participants are asked to fill out
survey questionnaires at the session. In the fourth model, the two methodologies are used
equally and in parallel to cross-validate and build upon each other's results. Social
marketers may operate under one or more of these models; the approaches are not
mutually exclusive.

A social marketing model for integrating methods must include quantitative and
qualitative methods at each stage of the process for formative research, process
evaluation and outcome evaluation. While each program is unique, the model proposed
here can be adapted based on available resources. See Figure 1, Integrative Social
Marketing Research Model.

Integrating Formative Research

During the formative research stage, in which the goal is to learn as much as possible
about how the target audience thinks and behaves in relation to the issue being addressed,
a host of research methods provides many different data "viewpoints" for seeing the big
picture. Exploratory research conducted at the beginning of the project reviews previous
research involving both quantitative and qualitative data and can include interviews with
those who have previously attempted to address the issue. This research will help in the
initial development of the project strategy to delineate the parameters of the project, steer
the selection of the target audience, specify the potential behaviors to be promoted and
identify lessons learned and potential pitfalls. Focus groups conducted for exploration
also yield valuable qualitative data regarding the target audience, providing insights into
their language, issues and obstacles they identify, and meanings attributed to beliefs and
behaviors.

Information learned from the initial focus groups can then be used to inform
questionnaire construction for a population survey to collect hard numbers for baseline
data. The survey will also help to segment the target audience based upon its distribution
across the stages of behavior change, as described by the Transtheoretical Model of
Behavior Change (Prochaska and DiClemente, 1983), or other characteristics. In addition,
commercial marketing databases, while quantitative in nature, provide highly detailed
profiles of target audience segments for message development and channel selection.

The messages and materials developed based upon the exploratory research should be
pretested using both qualitative and quantitative methods so that the results provide depth
of understanding as well as generalizability. Focus groups provide a valuable means to
pretest messages and materials, for audience members can provide spontaneous reactions
and explain their responses. This method, however, can only indicate trends and cannot
yield hard quantitative data needed for definitive decision making. If enough focus
groups are conducted and participants are considered representative of the target
audience, a survey questionnaire may be administered either before or after the focus
group to collect numerical data as well.

A central-site intercept survey, in which potential audience members are approached in a

public area and asked to respond to a quick questionnaire, provides another method of
pretesting materials. The fast turnaround nature of this method and high volume of
responses makes it ideal for testing draft executions of materials such as print or
television ads prior to production and implementation. This method is considered semi-
quantitative because respondents are not selected from a random sample, but questions
are usually closed-ended and tabulated statistically. Final decisions, such as choosing
from among several possible ads, can be made based on the numbers this method yields.

Integrating Process Evaluation

Upon implementation of the program, process evaluation helps to keep the project on
track and signals when changes are needed in the program strategy. The most common
data collection activity in this phase involves counting--materials distributed, number of
people attending activities, broadcasts of the television or radio ads, media coverage of
events, phone calls to the organization--to ensure that the project proceeds as intended.
Other quantitative tracking mechanisms, such as consumer surveys, identify whether the
program's message is reaching the target audience and is getting its attention and
motivating action. In an ongoing multi-year project, this may be a repetition of the
population survey conducted at the beginning; for a shorter-term project, a survey may
target a very specific audience segment.

Qualitative process evaluation methods can include periodic interviews or focus groups
with target audience members to assess their progress toward behavior change. Through
these activities, participants may inform program administrators of unforeseen barriers or
opportunities to adopting the behavior that need to be addressed to increase chances of
success. Observations of audience members may also provide clues to needed changes in
program strategy or messages in case they are using the product in an unsafe manner or
performing the target behavior incorrectly. The quantitative and qualitative process
research can be conducted simultaneously to collect and react to data.

Integrating Outcome Evaluation

Both types of research are instructive in identifying the program outcomes. A repeat of
the quantitative population survey will provide an indication of whether the program
realized its objectives in raising awareness, changing attitudes and initiating behavior
change. Related decreases in morbidity and mortality or other major indices will be more
difficult to claim without also conducting a matched community intervention study, with
the only difference between the communities being the presence of the social marketing
program.

In the end, the quantitative data emerging from the survey are generally used as the final
arbiters of success. However, qualitative research can point out successes that may have
occurred on a more human scale through anecdotes about how the social marketing
program made a difference in someone's life. Focus groups, interviews and other methods
of collecting individual people's stories and responses to the campaign are valuable in
learning which components of the program were successful and how the next project can
be improved. Both types of research are necessary to assess the full extent of the
program's impact upon the target audience.

Conclusion

Integrating quantitative and qualitative research methods lends depth and clarity to social
marketing programs. This combination of approaches is necessary because of the wide
range of data needed to develop effective communications. However, the potential for
problems exists when attempting to combine such divergent research paradigms; one may
end up not doing either type of research well. This integrative approach therefore requires
a research team with expertise in both types of methods. Using multiple approaches can
also be time-consuming, labor-intensive and expensive. Another obstacle, which will
likely change as social marketing gains in usage, is that combining multiple methods is
still not widely accepted as a viable research strategy--at least in mainstream public
health circles. As social marketers demonstrate that such research is necessary to fully
understand and address many health-related issues, the research norms and scientific
dogma regarding appropriate methods may shift to a new, more integrative paradigm.

The Limitations of Quantitative Methods for Analyzing Gerrymanders: Indicia,

Algorithms, Statistics and Revealed Preference

Micah Altman
Harvard University

Michael P. McDonald
George Mason University - Government and Politics; Brookings Institution

July 4, 2007

Abstract:
Simply put, all current quantitative methods are deeply flawed: Threshold rules based on indicia
that are hypothesized to be correlated with gerrymanders, such as compactness, margins of
competition, and estimated electoral responsive, are at best effective only locally and at worst
literally impossible to satisfy. Automatic maximization rules using these indicia or other
automatable algorithms universally ignore the political context in which they are applied and thus
yield politically biased results despite the appearance of neutrality. The most sophisticated
methods, which use computationally-intensive sampling from real districting populations, avoid
these problems, but suffer from intractable computational issues and (often) from implausible
formulation of the "null" hypothesis.

We place evaluating intent as a motive behind a redistricting plan into a formal quantitative micro-
economic framework to evaluate existing and emerging methods, and find that these methods are
statistically flawed. In place of classical statistical tests, we formalize a method of revealed
preferences to probe intent by comparing aspects of plans that were feasible, but not selected.
This method has been used in an informal, ad-hoc, manner in redistricting cases, but is not well
documented and has never been rigorously analyzed. Our method has five advantages. First, it is
easily interpretable. Second, it can be applied using only the data available to the original
planners and does not require estimating the outcomes of hypothetical elections. Third, lacking
sophisticated optimization technology, the basic method can be applied using hand drawn maps.
Fourth, it is more consistent with the knowledge that distracters had than statistical methods
because it does not implicitly assume that a districting authority was aware of all possible plans.
Finally, it is the only quantitative method for determining intent, so far proposed, that is
statistically sound.

Keywords: redistricting, gerrymander, competitiveness, compactness, hypotheis tests, bias,

responsiveness, automated redistricting

Problems in Quantitative Research

I It is necessary, first, to distinguish between two related but independent goals of sociology. The
first is social research, trying to get good descriptions of society and what is happening. The
second
is the problem of understanding society and of explaining what happens, e.g., the
effect of racism or why the crime rate has increased. We need the first to do the second, but the
first can be done without trying to do the second. Moreover, there are two quite different
approaches to social research. One can use a qualitative approach or a quantitative approach. In
what follows we talk only about quantitative social research.
II Roughly quantitative research is any sort of research in which something is counted or
measured—assigned a numerical value. It is critical to see that ‘data’ are not just given (the root
of
Latin datum is ‘given,’ and that in all instances ‘data collection’ is really the creation of data
where
none existed before. This usually goes unnoticed, perhaps because people have been doing this
for
centuries. For example, the first census was probably done in Rome. Once we have some data,
then, we can manipulate it. We can compute rates, e.g., the crime rate, or one can establish
patterns, e.g., trends like increases in the crime rate, or correlations between phenomena,
e.g., the correlation between poverty and crime, etc.
l. To count something one must clear about the unit of analysis. To discover that 48% of eligible
voters voted, we had to know that the unit was the eligible voter and we had to know how to
identify him/her. Consider, however, families, lower class youths, nursing staffs, small
businesses,
farmers, and in Hawaii, local, or Portuguese! Each of these raise conceptual problems which need
to be settled and there is no straightforward uncontestable way to do this.
2. Similar remarks apply to measurement. Anything can be measured. All (!) one needs to do is to
generate some criterion, what is called an indicator, for the variable being measured. Thus,
religiosity might be conceptualized along several dimensions, e.g., degree of participation in
ritual
behavior, degree of belief, etc. and then operationalized, given a number by means of a
questionaire: How times do you go to church? Do you believe in reincarnation, etc. As is plain,
there are genuine difficulties in knowing whether one has really captured the phenomena with
one’s
indicators. This is referred to as validity: Does the measure measure what it is supposed to
measure.
IQ (the intelligence quotient) is perhaps the most famous contested. But even if we have some
confidence in our measure, we can also ask whether it is reliable: Can it be counted on to measure
what it says it measures time after time?
3. When we combine several indicators into a single measure, we have an index, e.g., the
Consumer
Price Index, the Dow Jones Average, social class (measured as composite, e.g., of education,
income and occupation), Cost-of- Living-Allowances (COLA).
4. When we divide a measure by a measure, we have a rate, the number of crimes per thousand,
the number of unemployed divided by the number of those seeking employment, etc.
5. May I assume that we all can distinguish between a mode (the most frequent attribute), a
median
(the middle attribute in a ranked distribution) and a mean?
6. But in addition to computing rates and ‘averages,’ we can do more sophisticated forms of data
analysis, including, e.g., measures of dispersion (e.g., the standard deviation), measures of
variance,
and other even more sophisticated techniques. (It is not our purpose in this course to develop
these
or the various statistical techniques in wide use today. Sociology 275, a required course for all
sociology majors does this and more.)
But it is important, nonetheless, not to be mislead by what these methods do. Assuming that the
measurements are valid and reliable (above), all too often it is supposed (even by very
sophisticated
practicioners) that some result explains something when it does not. That is, it is easy to collapse
the
first goal (above, I) into the second. One example must here suffice.
In his fine textbook, Earl Babbie (Social Research for Consumers: Wadsworth, 1982) offers an
illustration. He constructs a fictitious example in it seems that there is a relationship between the
number of storks in different areas and the birthrates in those areas. (See his Figure 11.2, I)
attached). He writes that ‘this empirical relationship might lead one to assume that the number of
storks affects the birthrate.’ But, he continues, ‘an antecedent test explains away this relationship.
Rural areas have both more storks and higher birthrates than urban areas. Within rural areas, there
is no relationship between the number of storks and the birthrate; nor is there a relationship
between
urban areas’(p. 305). See Figure 11.2, II.
No doubt the incidence of storks does not explain the higher birthrate.
But we know this because we know that storks are not causes of births, and we know this because
we understand human reproduction—not because it is ‘explained’ away by appeal to the
urban/rural
variable. But we must emphasize his last sentence: ‘In comparison to the logical diagram
presented
to represent interpretation [figure 11.2], Figure 11.3 represents an explanation’ (my emphasis).
Presumably, given that empirically ‘all the rural places have high birthrates,’ the ‘Rural/Urban’
variable explains the birthrate. This is hardly the case, as a moments thought will confirm. The
birthrate is the aggregated result of countless decisions by couples. What we have is a pattern
which
needs explaining: Our analysis has revealed a pattern. The question now is, why do people who
live
in rural areas tend to make decisions to have large families. To answer this, we need to say some
other things, things which involve a considerable amount of explanatory theory.
For class use
Fall, 1989

Forecasting Techniques, Part 1:

Quantitative Methods
By Neil Mason, ClickZ, Jul 25, 2006
Sponsored by Omniture

Over the past few weeks, I've been taking a look at various analytical techniques that may
be appropriate for understanding more about visitor behavior than you may find in your
average Web analytics tool. Many of these techniques, like classification and
segmentation, involve the use of statistical analysis tools. This week, I'll continue in that
vein by looking at forecasting and some of the techniques used to assess and understand
future trends.

As businesses build their data trends, those trends become more interesting and useful.
One problem with a fast-growing environment in which all charts show lines shooting up
and to the right is it's difficult to know what the underlying trends are and whether
marketing activity is affecting this growth.

There are two broad categories of forecasting techniques: quantitative methods and
qualitative methods. Quantitative methods are based on algorithms of varying
complexity, while qualitative methods are based on educated guessing. I'll focus on
quantitative methods here. In part two, I'll look at qualitative methods.

Quantitative methods come in two main types: time-series methods and explanatory
methods. Time-series methods make forecasts based purely on historical patterns in the
data. Say you want to forecast site visitors over the next few weeks. Time-series methods
only use historical site visit data to make that forecast.

Explanatory methods use other data as inputs into the forecasting data. In the previous
example, you might include marketing data as inputs into a model to understand how
they affect visit levels and to forecast future visits with those data. These types of
techniques have been used for ages in the offline world to evaluate marketing activity's
effect on brand awareness or sales.

Time-series methods are probably the simplest methods to deploy and can be quite
accurate, particularly over the short term. Most quantitative forecasting methods try to
explain patterns in historical data as a means of using those patterns to forecast future
patterns.
Simple time-series methods include moving average models. In this case, the forecast is
the average of the last "x" number of observations, where "x" is some suitable number. If
you're forecasting monthly sales data, you might use a 12-month moving average, where
the forecast for the next month is the average over the past year.

Trouble is, simple averaging methods don't tend to work well when there's either a trend
in the data or seasonal effects. This tends to be the case in a lot of marketing data! In that
case, other techniques, such as exponential smoothing, may be more appropriate.

With moving averages, every data point carries equal weight in making the forecast. With
smoothing methods, more importance is placed on the most recent data than on the
historical data. If there's a trend in the data, it'll use the recent observations to make up
the bulk of the forecast, and the forecast is more likely to reflect the trend.

Moving averages and simple exponential smoothing techniques are available in Excel
and easy to execute. That's part of the great advantage of time-series methods: they're
generally simple, cheap to run, and relatively easy to interpret.

There are more complex time-series techniques as well, such as ARIMA (define) and
Box-Jenkins (define) models. These are heavier duty statistical routines that can cope
with data with trends and the seasonality in them. You'd probably need to invest in a
statistical analysis package or a dedicated forecasting package to use these more powerful
algorithms. Like any analytical technique, though, you shouldn't use them blindly or treat
results as gospel. All forecasts are invariably wrong, in fact. It's just a question of how
wrong they are.

So why would you use these heavier duty forecasting techniques?

Forecasting techniques are often used as much for their explanatory power as for their
predictive power. Understanding the trends and seasonal behavior of your business
provides a better understanding of its underlying health.

In consumer goods marketing, for example, these types of forecasting techniques are
often used to assess a brand's baseline performance. A forecast is made of what the sales
would have been in the absence of certain types of promotions or advertising so
underlying trends can be assessed.

Explanatory forecasting methods take the process a step further and allow you to relate
changes in marketing activity to changes in such outputs as sales, brand awareness, and
registrations. Here, we're looking for causality and can feed that into forecasts as a way of
evaluating marketing response. We'll take a look at this in more detail next time.

Till then...