Food Protection Trends - January/February 2022 - 10

Sampling and sample size
Food handlers that fulfilled the following selection
criteria were recruited for the study: (1) 18 years of age or
older, (2) directly involved in food handling, and (3) able
to communicate in either the English, Malay, or Indonesian
language. To encourage participation, the study provided
a monetary incentive and complimentary hand washing
facilities.
The sample size for this study was calculated using
G*Power version 3.1.9.2 software. Before sample size
calculation, the effect size was first determined using Cohen's
d effect size formula. M1
and M2
were the means for the
treatment and control group, respectively, obtained from
previous literature (61). The calculated Cohen's d effect size
was 0.829.
Cohen's d effect size = M1
- M2
SDpooled
where M1 is the mean for the treatment group (52.16), M2
is the mean for the control group (31.50), and SDpooled
/N1
= 1, and Cohen's d effect size = 0.829, a minimum
is the
pooled standard deviation (24.92) (61).
Based on a sample size calculation using G*Power version
3.1.9.2 software, where α = 0.05, Power = 0.8, allocation ratio
N2
sample size of 48 respondents (24 per group) was obtained.
Factoring in an additional 20% of the calculated value to
account for possible attrition rate (3), the final minimum
sample size was 60 (30 per group).
A total of 66 food handlers from 10 nationally funded
schools in Greater Kuala Lumpur, Malaysia, were recruited
using the convenience sampling method. However, five
participants withdrew during the study. In the end, 61
participants (treatment group [n = 31]; control group
[n = 30]) completed the program. To minimize sample
contamination, participants from control and treatment
groups were from different districts, A and B, respectively.
Sample contamination refers to sharing of information
related to the intervention among participants in the control
group of an intervention study (24). Selecting participants
from different districts may reduce the chance that they
would discuss the study with each other, which could
affect the outcome of the intervention. Between-group
comparisons were made to ensure that the participants
from the districts were similar.
Intervention
The intervention was implemented for 14 days prior to
the posttest. The treatment group was provided with the
following:
On-site interactive training
A 2-h on-site training using an iPad digital device (Apple,
Cupertino, CA) and informational posters was conducted
10
Food Protection Trends January/February
= 0.829
in each school by the main researcher assisted by a data
enumerator. The main researcher, a certified food handler
training instructor, conducted the training consistently
in each school in a treatment group. The data enumerator
assisted in the setting up and managing of teaching tools
during each session. Small-group training (four to nine
participants per session) was conducted in each school to
ensure training effectiveness. To keep participants engaged,
iPad digital devices (one to two participants per iPad),
posters, hands-on demonstrations, and activities were used.
Each session was conducted at the students' dining area
during nonpeak hours.
Using an iPad, the trainer gave a PowerPoint presentation
introducing five topics that incorporated some important
concepts from the literature (Table 1): (i) real case stories
(59), (ii) the use of disgust (45), and (iii) error-based
training (59). In topic 1, the training started with story telling
accompanied by a video clip of an actual food poisoning
incident that led to the death of a 5-year-old child in Kuala
Terengganu, Malaysia (5). Telling about a real case is
much more effective than showing statistics on foodborne
outbreaks because people are more likely to feel compassion
(59). For topic 2 there were slides showing sources of germs
and the microbial plate counts from different locations,
e.g., hands, tabletop, toilet, etc. In topic 3, " how germs are
transmitted, " images of feces and fecal-oral routes were shown
to create the feeling of disgust among participants (45). The
training was strengthened by the use of GloGerm powder
(Marlatek, Inc., Brockville, ON, Canada) and a handheld
UV light to demonstrate how germs can transfer from
different places due to cross-contamination and inadequate
hand washing practices. The last two topics (4 and 5) used
error-based training to cover contamination prevention and
Food Protection Trends - January/February 2022

Table of Contents for the Digital Edition of Food Protection Trends - January/February 2022
