2.3. Analysis
This study adopted a “mixed-methods” approach.
Quantitative data were entered in Statistical Package for the
Social Sciences (SPSS) Version 20®. The outcome variables included
the total number, type, and timing of recommended changes that
were implemented, and the number and type of “additional”
changes implemented by the organisations. The changes implemented
were dichotomised into higher order or lower order solutions.
Higher order solutions included elimination or engineering
strategies. Lower order solutions included such actions as the
provision of training programs and the introduction of job rotation.
The timing of the introduced changes was dichotomised into either
“less than 6 months” or “6e12 months” following receipt of the
report.
Initial, bivariate analyses using independent samples t-tests
were conducted for each outcome variable to identify an intervention
effect. Despite the randomisation of companies to the
standard and tailored approaches, there remained potentially
important differences in the characteristics of the companies in the
two groups. Poisson regression models were fitted for each
outcome variable, including a binary intervention variable (i.e.
Standard or Tailored), and variables representing workload, company
size, manager age/experience and budgetary control of the
manager to control for potential confounding factors. In the final
models statistical significance was defined as a two-tailed p-value
of 0.05 or less. Group allocation (i.e. Standard or Tailored) was
included in all the final models. The resultant incident rate ratios
(IRR) are interpreted as the ratio of uptake in the tailored group
relative to the standard group.
This study also adopted a thematic analysis of the qualitative
data obtained within a realist epistemological position. Whilst
the coding of the data was conducted inductively this was
focused at what has been described as the semantic level (Braun
and Clarke, 2006) across the whole data set with the transcribed
responses of each interview being read, re-read and coded openly
and independently by two members of the research team. Open
coding was conducted through identifying and labelling each
discrete incident, idea or concept expressed during the interview.
Where labels coincided, this allowed comparison between transcripts.
Conceptual labels were grouped together to develop
themes which were compared between researchers. Definitions
Table 1
Descriptive statistics of industry sector, nature of work and organisation size, according to the Australian and New Zealand Standard Industrial Classification (ANZSIC) and the
Australian Bureau of Statistics (ABS) and manager characteristics, including age, experience and control of OHS budget.
Standard/Tailored
Industry sector Nature of work Workgroup
size (n)
Workloada Organisation
size
Manager
Age
(years)
Experience
(years)
Control of
OHS budget
Standard Healthcare/Social Assistance Nursing 10 M Large 57 0.66 x
Standard Professional, Scientific, Technical Technical services 15 L Medium 38 4 x
Standard Public Administration & Safety Outdoor maintenance 19 M Large 54 20 √
Standard Manufacturing Sheet steel 14 M Large 46 3 x
Standard Healthcare/Social Assistance Charity collection 11 M Large 41 3 √
Standard Manufacturing Foundry 10 L Large 50 11 x
Standard Mining Call centre 18 S Large 53 8 x
Standard Mining Warehouse 10 M Large 44 3 √
Standard Healthcare/Social Assistance Disability services 14 M Medium 47 0.33 √
Standard Manufacturing Air-cond. maker 12 M Large 48 5 x
Standard Manufacturing Snack foods 9b L Large 32 1.8 √
Standard Manufacturing Snack foods 9b L Large 32 1.8 √
Standard Retail Retail 14 L Large 57 10 x
Tailored Other Services Laundry 20 M Large 61 2.5 √
Tailored Manufacturing Food manufacturer 19 M Large 36 2 √
Tailored Healthcare/Social Assistance Hospital orderlies 12 M Large 57 0.66 x
Tailored Healthcare/Social Assistance Hospital services 15 M Large 52 1.5 x
Tailored Public Administration & Safety Outdoor maintenance 16 M Med. 59 2.5 x
Tailored Public Administration & Safety Library 15 S Medium 59 2.5 x
Tailored Manufacturing Optical lenses 13 L Medium 35 2 √
Tailored Other Services Transport 19 M Large 30 2.5 √
Tailored Manufacturing Packaging 15 M Large 44 2 √
Tailored Retail Retail 13 L Large 62 4 √
Tailored Manufacturing Foundry 8b M Large 50 11 x
Tailored Professional, Scientific, Technical Technical Services 13 S Medium 38 3 x
a Workload was categorized, according to the Dictionary of Occupational Titles, into Sedentary (S), Light (L) or Medium (M). b Where the workgroup size is less than 10 this is due to the unavailability of individual workers during the company visits.
372 P. Rothmore et al. / Applied Ergonomics 51 (2015) 370e376
were discussed in relation to the original data and refined. Relationships
between themes were also explored, discussed and
interpreted.