4.2.4 Step 4: Conduct secondary research
Secondary data was obtained from various sources such as
journals, books, newspapers and Internet information.
Secondary data was used as part of section 1.7 of Chapter
One, Two and Three. All this data was utilised to supplement the findings of
the research.
4.2.5 Step 5: Select a primary research method
Primary data may be gathered through experimentation,
observations, interviews and written communication. The type of research will
eventually influence the choice of data collection. The method may comprise the
use of self-administered questionnaires, personal and telephonic interviews
that are mainly quantitative by nature (Du Plessis, 2010:118).
4.2.5.1 Data gathering technique
The following section below provides the motives for using
self-administered questionnaires to collect primary data. A research can make
use of different types of surveys such as personal and executive interviews,
mail and telephone surveys. Elements to take into account when choosing a
survey method are (Holder, 2008:72):
· The characteristics of the format in which data is
gathered;
· The quantity and quality of data which can be gathered
using a particular method;
· Control over the sample that differs according to
specific categories of surveys;
· The reliability of data which can be gathered with the
survey type;
· The speed of the research;
· The cost of the selected type of survey.
76
The characteristics of the format in which data is collected
are adapted from Holder (2008:123-125) and applied to the current research. The
reliability of data in the current study is assessed through pre-testing the
questionnaire using five patients who have received medical services at the
military hospital in Libreville, in Gabon. This was performed in order to
determine the time it takes respondents to complete the whole questionnaire,
and to spot and correct any problem in the structure of the questionnaire. The
self-administered questionnaires are handed out directly to the respondents of
the research. This makes the questionnaires easy to administer to all 200
patients and less costly to gather data from respondents compared to other
types of surveys techniques.
4.2.5.2 Self-administered questionnaires
Self-administered questionnaires refer to a survey method
that enables respondents to fill in a questionnaire. Survey questionnaires can
be issued via e-mails, fax, newspapers, Internet or through the place where
service is acquired such as a hotel, restaurant and hospital. They can also be
handed in personally as part of an intercept survey. Depending on the way the
survey is administered, there are a number of sampling framework issues such as
who can or cannot be contacted by fax or Internet, or whether there is a sample
bias. Benefits of a self-administered survey include anonymity of the
respondent in the research that can lead to the acquisition of more honest
answers. The questionnaire can also be completed at the convenience of the
respondent. Since an interviewer is not required, errors or bias due to
interviews is eradicated. The cost of accessing a geographically scattered
sample is cheaper for most types of self-administered surveys than personal or
telephonic surveys. In the majority of self-administered surveys, there is no
influence over who is completing the questionnaire (The Self-Administered
Survey, 2012).
In this study, self-administered questionnaires were handed
to respondents to complete. The respondents in the study had to be 18 years or
older, males and females, who have experienced medical services and stayed over
at the military hospital for at least one night. Data was collected over a
period of three weeks.
77
4.2.5.3 Covering letter
The cover letter informs respondents about the goal of the
research, it should:
· Encourage respondents to be involved in the study;
· Provide the reasons why respondents have been chosen;
and
· Describe the benefits of the research to respondents.
Information provided in the covering letter should influence
the respondent's willingness to be involved in the study (Du Plessis,
2010:120). A covering letter was generated by the researcher. It described the
goal of the research and guaranteed the confidentiality of all respondents'
details and information. A cover letter was issued upon the respondent
request.
4.2.5.4 Questionnaire
The questionnaire included self-administered items initially
developed in English and translated into French, the main language spoken in
the study location. Each translation was verified by professional English and
French translators to ensure that the translations were correct. It is the
major research instrument used for the investigation. A questionnaire
represents an instrument that gathers primary data in survey-based research. It
is composed of well-structured written questions and fixed responses
alternatives directly linked to the purpose of the study (Van Vuuren, 2011:93).
Unlike other forms of surveys, in self-administered questionnaires, researchers
do not actually need to be present during the interview. However, they can
distribute the questionnaires to respondents who are free to read and complete
them at ease (Zikmund & Babin, 2007:143). The benefits of using this
instrument are as follows:
· The researcher can gather all completed questionnaires
within a particular point in time.
· Any question or point that is not clear to respondents
can be explained on the spot.
· The researcher can have the chance to bring in the
research topic and encourage respondents to provide their honest responses
(Munyaradzi, 2010: 215).
· Distributing questionnaires to a large group of people
at the same time appears to be less costly and time consuming than face-to-face
interviews.
78
? However, self-administered questionnaires often require well
through and clarity on the written questions in order for respondents to easily
understand the questionnaires (Zikmund & Babin, 2007:143).
The survey questionnaire is divided into five major sections:
Section A which consists of questions about the respondent's general
demographic information, section B that consists of questions related to the
respondent's perceptions of services, section C that consists of the tangible
aspects of the hospital, section D that relates to the overall perceptions of
services quality, and section E which consists of a list of factors that could
influence the decision to choose the military hospital. The questionnaire
contains 50 questions in total. Section B includes16 statements that measure
the items of service quality based on the reliability, responsiveness,
assurance and empathy dimension. Section B is in the format of a seven-point
Likert-type scale ranging from 1 (strongly disagree) to 7 (strongly agree)
where respondents were given specific limited-alternative responses and asked
to select the one closest to their views. A seven-point Likert-type scale is
best used when the items in the questionnaire are related to each other, or
when the options are in the form of a scale. This type of scale is suitable
when a research aims at getting an overall measurement of a particular topic,
opinion, or experience and also to gather particular data on contributing
factors. Measuring the satisfaction level of a recent service experience is a
common use. (Likert Scale, 2012).
The research questionnaire was adapted from the questionnaire
used by Holder (2008:124) who previously developed and tested items related to
service quality in the hospital industry for their validity and reliability.
4.2.5.5 Questionnaire design
Questionnaires are usually built to capitalise on the
importance and precision of data collected, to capitalise on the involvement of
all respondents in the study, and to facilitate data gathering and their
evaluations. The reasons researchers often use a seven-point Likert-scale are
that they lower the development of response bias among respondents, and they
assess behaviours, beliefs, views and perceptions. Likert-scale responses are
usually not difficult to code and to analyse straight from questionnaires (Du
Plessis, 2010:121).
79
Section A of the questionnaire consists of
eight questions related to the socio-demographic details of the respondents.
This enables the researcher to assess the socio-demographic background of
patients attending the military hospital in Libreville, in Gabon to know if
variables such as age, education, and marital status can influence how patients
rate doctors and nurses on the service quality dimensions in the military
hospital in Libreville, in Gabon. These questions include the following:
1. Gender (Sèxe)
2. Age (Age)
3. Highest educational level (Niveau d'
éducation)
4. Marital status (Statue conjugal)
5. Employment level (Forme d' emploi)
6. Residential area (Lieu de résidence)
7. First time patient? (Est-ce votre première visite
dans cet hôpital?)
8. Were you required to stay over for at least one night?
(Etes-vous obligé de passer au moins une nuit dans cet
hôpital?)
Section B consists of sixteen different
statements on experience of service quality based on the dimensions of service
quality such as reliability, responsiveness, assurance and empathy. The
objective is to compare the views of patients regarding the service delivery by
doctors and nurses at the military hospital. The reliability and the empathy
dimension had five statements each: the responsiveness, the assurance and
tangibility had four statements each. These aspects were adapted from Holder
(2008:124). They are discussed next.
(a) Reliability
Items measuring the influence of reliability on satisfaction
were adapted from Holder (2008:124) who previously developed and tested items
related to the influence of reliability on satisfaction in the hospital
industry. Statements related to reliability include items 1 to 4.
80
The statements read as follows:
The medical doctors/nurses who treated me...:
(Je crois que les médecins /infirmières qui
m'ont traité...)
1. Acted in a way that caused me to trust them (ont
suscité en moi une confiance en eux/elles)
2. Acted in my best interests (ont agi dans mes
intérêts personnels)
3. Had the ability to examine me properly (avaient la
capacité de m'examiné correctement)
4. Were always honest with me (ont été
honnêtes envers moi)
(b) Responsiveness
Items pertaining to the influence of responsiveness on
satisfaction were adapted from Holder (2008:124) who previously developed and
tested items related to the influence of responsiveness on satisfaction in the
hospital industry. Statements related to responsiveness include items 1 to
4.
The statements read as follows:
The medical doctors/nurses who treated me...:
(Je crois que les médecins /infirmières qui
m'ont traité...)
1. Informed me of my state of health during consultation
(m'ont informé sur mon état de santé lors de la
consultation)
2. Communicated the prescription of medications for my treatment
with me in a manner that I could understand (ont utilisé un langage
qui m'a permis de comprendre les prescriptions médicales pour mon
traitement)
3. Always responded to my queries (répondaient
toujours à mes questions)
4. Always listened to what I had to say (faisaient
toujours attention à ce que j'avais à dire)
81
(c) Assurance
Items pertaining to the influence of assurance on satisfaction
were adapted from Holder (2008:124), which include items 1 to 4.
The statements read as follow:
The medical doctors/nurses who treated me...:
(Je crois que les médecins /infirmières qui
m'ont traité...)
1. Showed respect and dignity towards me (m'ont
traité avec beaucoup de respect)
2. Carried out their tasks competently(m'ont traité
avec beaucoup de compétence)
3. Respected the confidentiality of my treatment (ont
respecté la confidentialité de mon traitement)
4. Were well trained and qualified (avaient beaucoup d'
expertise)
(d) Empathy
Items related to the influence of empathy on satisfaction were
adapted from Holder (2008:124), which include items 1 to 4.
The statements read as follow:
The medical doctors/nurses who treated me...:
(Je crois que les médecins /infirmières qui
m'ont traité...)
1. Provided me with individual attention (m'ont
accordé une attention particulière)
2. Were concerned about my well-being (ont montré
beaucoup d'intérêts pour mon bien-être)
3. Understood my specific needs (ont compris mes besoins
personnels
4. Cared about me (ont pris soin de moi)
Section C consists of five different
statements on tangibility and items related to the influence of tangibility on
perceptions of services which were adapted from Holder (2008:124). The
objective is to establish the perception of patients regarding the tangible
aspects of the military hospital in Libreville, Gabon. The statements on
tangibility read as follow:
82
Thinking of the military hospital, I believe.... (Je crois
que...)
1. The furniture in the military hospital such as beds and
chairs was comfortable (le mobilier de l'hôpital tel que les lits et
chaises est confortable)
2. The interior decoration in the military hospital was
aesthetically appealing (J'ai aimé le décor interne de
l'hôpital)
3. The military hospital had a pleasant atmosphere (J'ai
aimé la sensation atmosphérique de l'hôpital)
4. The colour scheme at the military hospital was attractive
(les couleurs de l'hôpital étaient attirantes)
5. The toilet facilities in the military hospital were clean
(Les toilettes étaient propres)
Section D consists of five different statements
on the overall perceptions of the services received from the military hospital.
The objective was to assess how patients perceived the overall services
received from the military hospital.
The statements on overall perceptions of the services received
from the military hospital read as follow:
Thinking of the military hospital, I believe.... (Je crois
que...)
1. The overall quality of the administration service at the
military hospital was excellent (le service administratif de
l'hôpital était excellent)
2. The overall service provided by the military hospital was of
a high standard (le service offert par l'hôpital était en
général de bonne qualité)
3. The overall service I received at the military hospital met
my expectations (les services reçus de l'hôpital se
rapprochent de ce dont j'espérais recevoir)
4. I felt good about coming to this military hospital for my
treatment (je me suis bien senti(e) dans cet hôpital)
5. Overall, the service offering of this hospital was superior
to the one offered by government hospitals (les services offerts dans cet
hôpital étaient meilleurs que ceux offerts par les hôpitaux
publiques)
83
Section E consists of a list of sixteen
factors that could influence the decision to select the military hospital.
These factors were adapted from Doghaither, Abdelrhman, Saeed and Magzoub
(2003:107). The objective was to establish which factor could influence
patients' decision to choose the military hospital in Libreville, Gabon.
The statements read as follows:
1. Availability of modern medical equipment (La
présence d'un équipement médical moderne)
2. Specialised physicians (Les médecins
spécialisés)
3. Quality of nurses (La qualité des
infirmières)
4. Availability of medicines (La disponibilité des
médicaments)
5. Relatives living in the hospital area (Un membre familial
vivant à proximité de l'hôpital)
6. Hospital near residence (La proximité de
l'hôpital du lieu de résidence)
7. Ease of getting to hospital location (L'accès
facile à l'hôpital)
8. Care cost (Le coût des soins hospitalier)
9. Patient's past experience (L'expérience
passée du patient)
10. Competent staff (La compétence du personnel
médical
11. Know people who went to the hospital (Connaissiez-vous
des personnes ayant fréquenté cet hôpital auparavant
?)
12. Good reputation in media (la bonne renommée de
l'hôpital)
13. Hospital cleanliness (l'hôpital est
propre)
14. Employees are friendly towards patient (le personnel est
amical)
15. Hospital is well organised (l'hôpital est bien
organisé)
16. Other? (autres?)
4.2.5.6 Pretesting of the questionnaire
Pretesting a questionnaire is vital to ensure its validity and
reliability. It refers to testing the questionnaire using a small number of
respondents to spot and eliminate any possible error in the survey design
(Malhotra, Hall, Shaw & Oppenheim, 2008:233). Parasuraman, Grewal and
Krishnan (2007:303) stated that pretesting is crucial in a way prospective
respondents may help identifying any problem in the questionnaire. All
84
sections of the questionnaire such as question content,
sequence, wording, form layout and question complexity should actually be
pretested. Additionally, respondents of the small sample must have the same
traits as those included in the current study (Aaker et al., 2007:247). The
purpose of pretesting is to make sure that questions are clear and easy to
understand. Furthermore, pretesting also intends to ensure that questions are
extracting the answers required and that any misinterpreted questions are
eliminated before the questionnaire is distributed to a larger extent
(Munyaradzi, 2010: 222).
The questionnaire was pretested using five patients who had
received medical services at the military hospital in Libreville, Gabon.
Pretesting a questionnaire refers to using it on a trial basis. Pretesting is
important, since it enables knowing that the developed questionnaire is
adequate and will fulfil the purpose of the research at hand. Moreover, the
gathered data will be appropriate and as correct as possible, the target
respondents in the study will participate as fully as possible, and evaluation
of data will run smoothly. In this study, the relevance of pretesting was to
pinpoint flaws in the questionnaire, and to assess the time needed for a
respondent to fill in the entire questionnaire. Pretesting the questionnaire
was also used to examine its face and content validity, and to detect and
correct problem areas. During the pilot study, issues such as the difficulty of
respondents to understand the sentence structures and the difficulty to
understand questions were identified. These issues were addressed by obtaining
suggestions from respondents for revising questions and the structure of
sentences. Once the questionnaire had been pretested, it was then refined for
the data to be gathered (Roberts-Lombard, 2006:41).
4.2.5.7 Reliability and validity of the
questionnaire
Reliability and validity are important elements to evaluate
the trustworthiness of any study. Reliability refers to the extent to which a
scale provides the same outcomes if repeated. To determine the reliability of
the measurement scale in this study, the internal consistency reliability test
was used. It is regarded as an instrument utilised to test the reliability of a
measurement scale or homogeneity among the variables being measured (Malhotra,
2009:315; Churchill & Brown, 2007:269; Saunders, Lewis & Thornhill
(2007:367). The simplest measure of internal consistency is split-half
reliability. The variables on the scale are divided into two halves and the
resulting half scores are compared. High similarities between the halves show
high internal consistency
85
(Parasuraman et al., 2007:270). Additionally, internal
consistency is also used to evaluate the reliability of a summated Likert-scale
where several variables are summated to form a total score. To assess the
reliability of the homogeneity of the measurement scale, the Cronbach's alpha
was used for this research.
Cronbach's alpha is perceived as the average value of the
reliability coefficients one would have for all possible grouping of items when
split into two half-tests. The alpha value is the average of all possible
split-half coefficients that result from different splitting of the items in
the scale. The desired cut-off score for a measurement scale to be reliable is
0.6. Other scores suggest unsatisfactory internal consistency reliability
(Munyaradzi, 2010: 215).
Table 4.1 illustrates the Cronbach alpha values for the
different variables in the study at hand.
Table 4.1: Reliability statistics (Doctors and
Nurses)
|
CONSTRUCTS
|
|
|
Cronbach's Alpha
|
|
DOCTORS
|
NURSES
|
|
Reliability
|
|
|
0.809
|
0.767
|
|
Responsiveness
|
|
|
0.735
|
0.717
|
|
Assurance
|
|
|
0.650
|
0.780
|
|
Empathy
|
|
|
0.734
|
0.666
|
|
Cronbach's Alpha
|
|
|
Tangibility
|
|
0.600
|
|
Table 4.1 illustrates that Cronbach's alpha for all five
constructs is above the lower limit of acceptability, 0.60. This confirms the
reliability of the measurement set for this study. The validity of a scale is
the extent to which it is a true representation of the underlying variable it
is trying to evaluate. Validity is measured using content validity and
construct validity (Parasuraman et al., 2007:269). For the purpose of this
study, content validity was used to assess the accuracy of the measurement
scale.
Content validity can be evaluated by subjective agreement
between experts, that a scale logically seems to accurately measure what it is
intended to measure. Another way to ensure content validity is to use a panel
of people to assess how well an instrument meets the purpose of the study
(Malhotra et al., 2008). In the current study,
86
content validity was determined by people including
statisticians and academics with the expertise for the research at hand. The
questionnaire was evaluated and the necessary modifications were noted both in
terms of the measurement items and the measurement instrument as a whole.
Content validity was evaluated by determining the variables defined and used in
previous literature.
| 
