Research Design

Methodological Framework
The aim of this study is to examine the publics perspective on the recent
engagement of high school athletes as brand ambassadors and to assess
whether this perception has had any impact on their respective brands.
According to the Consortium on Cognitive Science Construction, no
single research method can answer all questions about the nature of
perception. It states that every method has limitations as well as some
foundational questions are more than empirical questions, rather there are
philosophical questions. As such, for the purpose of this study, a mixed
method approach of investigation will be chosen. A mixed method approach
involves integrating quantitative and qualitative data collection and analysis
in a single study or a program of enquiry. Creswell et al 2003. According
to Bulsara, the mixed method approach which is also called multimethodology or triangulation which evolved from evaluation research, is
useful whenever there are answers needed from a number of perspectives
and where one methodology does not provide all the information required. It
also ensures that pre existing assumptions from the researcher are less likely
and that there are no gaps to the information or data collected.
Examining the athletes perspective on the recent engagement of
interscholastic athletes will be obtained using qualitative research methods
in the form of interviews and focus groups. Given that this study is an unresearched phenomenon, the use of qualitative methods which are
exploratory in nature will seek to generate novel insights on the issues. The
qualitative methods will be used to generate data to gain a greater
understanding from the athletes themselves on their perception of being
engaged by corporate entities as brand ambassadors. This will guide the
decisions made by corporate entities on their approach as it relates to
corporate sponsorship of high school athletes. This approach will also be

used to develop and adjust the quantitative measurement processes and

instruments.
Quantitative research will be utilized to explore publics opinion with the aid
of surveys. Survey items will be administered via telephone interviews.
Administering the survey by telephone is considered the most efficient and
cost effective method for surveying the public.
download/reports/GA_Litter_Survey_Guide.pdf. The use of both of these
approaches will capitalize on the respective strengths of each approach. Jick
T. Mixing qualitative and quantitative methods: triangulation in
action. Adm Sci Q. 1979: 602610. The qualitative research methods will
be used to corroborate findings, generate a more complete data and using
the results to enhance insights attained with the quantitative method.
Creswell JW, Piano Clark V. Designing and Conducting Mixed
Methods Research. Thousand Oaks, Calif: Sage; 2007.
The face to face interviews and administering of survey will be done
sequentially. Preliminary qualitative interviews will serve to generate or
adjust the content for the questionnaire that will be used in the follow up
quantitative study. Or

Conversely, a preliminary quantitative component

may generate surprising or inconsistent findings that may be examined in

greater depth with a follow-up qualitative component The reason this is
chosen is that....
Sampling Strategy and Size
In order to obtain information about the publics perception, the population
that ill will be studies is the public in general. In survey research, the public is
often referred to as the general population. Reliable data can be obtained
from surveying a representative sample of the population and not
necessarily the entire population. According to
http://www.cios.org/readbook/rmcs/ch05.pdf, theories are formulated

from a subset of the population and the results used to generalize for the
entire population, since the hypothesis cannot be tested on the entire
population. Such a subset is the sample and it is considered representative
of the population and it must be selected in a way that any conclusions
drawn from it can be generalized.
For the purpose of this study, there will be two phase, the first phase is to
conduct the preliminary interviews and focus groups and the second phase is
the administering of the survey items. The size of the sample affects the
accuracy and reliability of the data you collect. In general, the larger the
sample size, the more reliable the data you collect will be and the more
accurately it will represent the opinions of the entire population.
http://www.responsivemanagement.com/download/reports/GA_Litter
_Survey_Guide.pdf
In determining our sample size, each Parish boundary and a population
estimate was taken into consideration based on the data obtained from the
2011 Population Cencus, conducted by Statin. statin ja.
http://statinja.gov.jm/Census/Census2011/. This was done to ensure that the
sample was representative and can be used to make generalizations about
the entire population, even though th entire population cannot be tested.
With a margin of error of plus or minus 10%, and a confidence level of 95%, a
sample of 708, 276 will be selected.
http://www.surveysystem.com/sscalc.htm#one. The total number will be
broken down across all 14 parishes to ensure a more representative
response.
The method of sampling that will be used is the stratified random sampling.
This method is a probability sampling technique and according to----, using a
probability technique will allow generalizations to be made about the
population. The population will be stratified according to region. In order to
correctly obtain a sample that will be representative of the population, a

sample will be obtained using telephone records from Lime and Digicel. More
than the required number of participants will be selected due to the
understanding that some records will be disconnected or the individual may
not want to participate in the survey. The participants for the focus group
and interviews which will be high school athletes will be selected based on
simple random sampling. A sampling frame will be requested from ISSA, and
a random number drawn.
Research Design
The study plan will involve the gathering of information on the engagement of high
school athletes as brand ambassadors from the general population. The preliminary
interviews and focus group will obtain information from high school athletes.

Data Collection procedures

The preliminary data will be collected by face to face interviews and focus groups.
When that phase is completed and if necessary the survey instrument is adjusted or
modified, the quantitative data will be collected by making survey calls. A pretest
will be conducted of the survey instrument to allow for final edits and corrections to
be applied to the entire survey administration processes. Once all pretests
conducted, all errors are identified and corrected, data collection will begin. This will
be carried out by out by our team of experts, who have been trained and equipped
with skills necessary to administer survey instruments via telephone which are
essential to a successful and scientific defensible survey. The survey data will be
collected and stored on the computer assisted telephone interviewing (CATI)
software, Questionnaire Programming Language(QPL) and the data analysis
software package that will be used is the Statistical package for Social Sciences
(SPSS).
Telephone calls will be made each day at different times of the day, in order to
include both employed and unemployed. Calls will be made until a response is
given, whether the acceptance or rejection is given to complete the survey. It is

important to keep non response to a minimum to avoid response bias. Even though,
as indicated by http://www.cios.org/readbook/rmcs/ch05.pdf, response bias is
difficult to avoid and it occurs when the cases chosen for inclusion in the sample
systematically exclude themselves from participation in the research. To overcome
or minimize non response rates we will be proposing incentives of cell phone cards
to persons who complete the survey instrument. In addition, making multiple calls
at different times will help maintain the representativeness of the sample and avoid
selection bias. In other words, avoid a bias towards people who are easy to reach by
telephone and thereby providing and equal opportunity for all to participate. Calls
will be made until the sample size goal for each parish is reached.

In assessing whether the corporate entities brand has been affected by

public perception as result of the engagement of high school athletes as
brand ambassadors, one instrument will be used to measure the
independent and dependent variable. The independent variable in this case
will be the companies brand, with the public perception being the
dependent variable.
Ten items on the instrument measures brand perception, with 4 being a
Likert Scale measure while the other items are measures of moderating
variables.
The reliability and validity of the instrument will be checked. In addition, both
internal and predictive validity of the instrument have been demonstrated to
be high. In addition to the above independent and dependent variables,
three secondary independent or moderator variables will be considered.
According to Tuckman (1978), a moderator variable is a factor which is
measured, manipulated, or selected by the researcher to discover whether it
modifies the relationship of the independent variable to an observed
phenomenon (p. 63). Thus, the moderator variables can determine the
extent to which the relationship between the two major variables is
influenced by secondary factors. In this study the moderator variables of
age, gender, and location will be included.

Limitations (POTENTIAL SOURCES OF ERROR AND SOLUTIONS)

There are two main types of potential sources of error: errors related to
sampling and errors
related to nonsampling. Errors related to sampling result from samples that
are not
representative of the population under study (for example, general
population telephone surveys
using telephone books or other lists) and the sampling error, intrinsic in all
sample surveys.
Nonsampling errors include poorly-worded questions, poor interviewer
techniques, and mistakes. in data compilation and analysis. Nonresponse
bias can be an important source of error. Nonresponse bias results when the
people who do not respond to a survey hold different attitudesor opinions. A
blatant nonresponse bias occurs when a general population telephone survey
is conducted during the daytime hours only, between 9 a.m. and 5 p.m.
Since older people are most likely to be home during the day, older
individuals are likely to be oversampled and thus overrepresented in the
results. This is why it is important to survey at all times of the day (9 a.m. to
9 p.m.), as well as on different days of the week. This also illustrates why the
demographics of the survey respondents should be compared to the
demographics of the population under study once the study has been
completed.
Nonresponse bias can be minimized using techniques to maximize response
to the survey. For
telephone surveys, such techniques include multiple call-backs and trained,
confident
interviewers with clear voices. Bias in results also can occur from respondent
fatigue. As the length of the interview or questionnaire increases, an
individual is likely to tire more, and the chance of making mistakes increases.

Surveys that are too long or ask many similar questions can result in a
respondent either hanging up or simply giving erroneous answers to finish
the interview as quickly as possible. This is another important reason to keep
surveys short.
Effective pretesting can alleviate many problems associated with response
bias. Before a survey
is conducted it must be pretested. The designers of the survey should listen
in on the pretest if it is a telephone survey and listen for respondent reaction
to each question. If respondents have difficulty with questions, do not
understand questions, or are clearly misinterpreting questions, the question
must be reworded. The interviewers administering the survey should be
queried on respondents reactions to the survey as a whole as well as each
individual question. Were the questions simple and easy to understand? Did
the survey logic work? Involving interviewers in
the analysis of a pretest is an important part of the pretest. If the survey is a
mail survey it should be pretested via mail, although focus groups can
sometimes be used to pretest survey
instruments. The researchers should carefully inspect the returned surveys.
Are answers
missing? Why? Is there a pattern to missing answers? Are there inappropriate
answers? Such
evaluations can identify any potential response bias.
Data Analysis
The data will be analyzed and interpreted with the use of SPSS