Respondents of the Study (

¶ … Respondents of the Study ( Population Characteristics)

The staff group was a convenience sample surveyed from five different inpatient wards that were located in local hospitals situated in (* a city in Australia?). There was a total of 100 participants, with * percentage of them being female, ranging from age * to *, the average age being.. The mean age of nurse's employment in the practice was * . Ethnicity ranged with * percent being Caucasian, * percent Blacks, and * of an alternate ethnicity.

Sampling and the Sample size

Factors such as accuracy and confidence, population size, time and cost constraints were taken into consideration in selecting sample size. Using the non-probability sampling technique where a convenience (accidental or haphazrd) sampling strategy was employed, a total of 100 respondents were selected as a sample of the study from five Mental Health inpatient wards ( this is the whole available population from these 5 wards). The respondents come from various job titles ( health care support workers to Ward Managers) and work experiences in order to enhance ethnic distribution and other differences so as to increase the generalization of the result.

Survey Instrument:

The survey was intended to assess the impact of the independent variable of Productive Mental Health Ward program on the dependent variable of staff satisfaction.

The questionnaire has two sections. The first section of the questionnaire invovled demographic questions (such as identification for pre and post comparison, work banding, optional sex and age, and length of experiences in the current ward and with NHS). The second section contained 21 questions under five categories where the staff can score their subjective perception from 0 to 5, where 0 is strongly disagree, 1 is moderately disagree, 2 is slightly disagree, 3 is slightly agree, 4 is moderately agree and 5 is strongly agree.

Category 1 assessed the working environment -- facilities, easy to find equipment and ward cleanliness.

Category 2 assessed the team work, staffing level and availability of information for them to work safely.

Category 3 assessed the ward routine -- workload, direct time with service users and provision of safe and supportive care.

Category 4 assessed ward management -- empowerment, management support, visible management, value etc.

Category 5 assessed staff well-being -- team support, stress, work-life balance and able to leave work behind after the shift.

The adjusted survey (see Index) contained the following additions: (a) staff identification, (b) staff grade or band, (c) staff age (d) sex of nurse (e) length of time in the current ward, and (f) length of service in NHS.

In all other ways, my survey stayed constant to the original including in that the subjects were asked to respond the same set of questions, in the same order, and that all the subjects had the same set of options for their responses

Instrument Development

It was suggested that the survey used currently as instrument would be employed to investigate impact of program and nurse's satisfaction with their work, and it was also decided to modify the questionnaire in various ways. Suggested recommendations on the part of the researcher were inclusion of the following: a) staff identification, (b) staff grade or band, (c) staff age (d) sex of nurse (e) length of time in the current ward, and (f) length of service in NHS.

We added these variables in order to establish the influence of various factors of this very subject in their job satisfaction.

Since this survey is part of the PMHW project, it did not require further permission to use this questionnaire for final phase wards. However, permission was solicited from the Service Improvement and Innovation Steering Group to implement changes. Since the altered questionnaire had not been validated or standardized, I performed the reliability test-retest to validate and standardize the questionnaire.

A pilot study was done for this purpose with the survey test on 20 participants who were selected from one of the first phase wards for the pre and post test analysis to validate the questionnaire. These 20 participants matched the larger population in demographics and profession. The survey was conducted at the end of April with a retest performed 3 weeks later.

The week following the training, I presented the staff with the adapted questionnaire *handing them to each one individually and retuning to each individual the following day for receipt of completed survey.

The data of the questionnaire was analyzed using SPSS, and, since it was dealing with ordinal data, Kendall's correlation coefficient.

As this is a service evaluation, I will be feeding results of this study back to our Executive Management Team, participants and other steering group. Results will be incoporated in the final evaluation, after which I will test the instument on a larger population the following week. I intend to assess the validity of the addended instrument by analyzing the pre and post results ( impact or changes in 6 months during the implementation of initiative) using SPSS with the Kendall's correlation The SPSS sheet will provide the validity and reliability of the questionnaire coefficient being the chosen method of data analysis.

I would expect the initiative to positively impact on the frontline staff's job satisfaction, empowerment to make changes, improve the work environment etc. The survey instrument will help me to measure the impact before and after the implementation of this project.

Response scale

The survey itself was rated according to a Likert scale of 0 to 5 where 0 represented 'Strongly Disagree; 1=Moderately Disagree', 2= Slightly Disagree; 3= Slightly Agree; 4=Moderately Agree; 5=Strongly Agree.

The survey was, further, divided into five categories. These were: Environment; Teamwork; Ward Routine; Ward Management; and Staff Well Being. Three items followed the first section, four items followed the next two sections, and the last two contained five items. Following each section was a space requesting comments or feedback regarding that section

Examples of questions included:

1. The staff facilities on my ward meet my needs (Environment)

2. We work well as a team (Teamwork)

3. I am happy with the workload that I have on each shift (Work Routine)

4. I feel empowered to make changes (Ward Management)

5. I feel supported in the team (Staff Well-being)

Validity and reliability of the instrument

The reliability of the instrument was assessed by means of the internal consistency and the test-rest coefficient. Using van Saane et al.'s (2003) study as basis, an instrument with an internal consistency coefficient of 0.80 or higher was considered adequate. In the absence of scale total, the ranges of the sub-scales in the instrument are considered. According to *Lloyd et al., the test-retest conflict, too, has to be 0.70 or higher. These are my measurements for reliability of the instrument.

Assessment of validity of the instrument is divided into three categories: (a) convergent validity, (b) discriminant validity, and (c) content validity. The last is addressed in the following section.

Convergent validity refers to the degree of correspondence between created instrument used in this study and a similar instrument employed for the same purpose (*17, 18). The degree of convergence assessed here would be correlation between this appended instrument and the original questionnaire on which it is based. Since it is closely connected to the original instrument, a moderate to high correlation is expected. Based on, the criterion of convergent validity was set for a minimal of 0.50, or the sub-scale correlation range starting from that same rate.

Discriminant validity is the reverse of convergent validity and defined as the degree of difference between created instrument used in this study and a similar instrument employed for the same purpose (*17, 18). There are three features of discriminant validity, two of which were not related here. The first is to investigate whether the instrument employed here is similar to another instrument, which, although similar, measures a different concept. This condition was irrelevant here, since the original instrument measures the same concept. Secondly, responsiveness between both instruments is investigated. This will best be known after reiterated testing of the modified instrument since an instatement first has to show reiterated stability (established by constant testing) before its responsiveness can be assessed.

Content Validity, Reliability

Content validity refers to the extent to which the instrument measures the concept defined by the researcher and that which the researcher intends to evaluate (*18). The content validity was assessed by examining the fit between the conditions associated with nurse satisfaction of work that were retrieved from the literature search. Since 'job satisfaction' is a tenuous and abstract conceptualization fluctuating from country to country and never definitely or unanimously assessed, I used the recommendation offered by the European Foundation for the Improvement of Living and Working Conditions (EFILWC). They advised that only a tenuous description can be gained and that research should take into account the country in which the original report was authored since the definition of 'job satisfaction' tends to hinge on national constructs. Since, as yet, no common framework for defining 'job satisfaction' exists, I employed the most recent and in-depth research on the subject as rough aid to the matter.

Sources of data

*Two literature searches were conducted in order to generate conditions of nurse satisfaction. One search centered on variables of work satisfaction described in relevant job satisfaction theories. The other search probed applicable reviews and meta-analysis for factors of job satisfaction relevant to the nurse profession.

The literature, in the first case, was plumbed for studies that provided theoretical insight into the concept of job satisfaction. Terms such as 'job satisfaction' and 'theory' were used to aid the research.

In the second case, research was conducted by using terms such as 'nursing', 'job satisfaction' and 'factors'. The resulting Meta analyses and literature reviews were thoroughly analyzed and only those relevant to the intent of the study retained. They had to, in other words, be congruent to the nursing profession and related to satisfaction that the employee gained from his or her job.

'Nurse satisfaction' as a result was, accordingly, broken down and defined into the following *ten categories: (1) Workload -- whether staff workload was reduced as a result of the program; (2) Empower -- whether staff felt greater empowerment and ability to do whir work; (3) Support -- whether staff felt that they received more support as a result of the program; (4) Opportunity: whether the program afforded them more opportunity; (5) Stress- whether implementation of program reduced usual levels of stress; (6) Work life- whether work life was enhanced as a result of the program; (7) Leave -- the number of nurses who left since the program was introduced was also evaluated (8) Valuable -- finally, the value of the program to the nurses was considered. (9) Team -- whether staff considered there to be a greater team spirit since program had been introduced (10) Information- whether communication / transmission of information had been enhanced as a result of the program.

Content validity would be estimated as 'adequate' according to the number of these factors incorporated in the final instrument. The fewer the number incorporated and tested, the lower would be the rating of content validity. I therefore aimed that at least three such factors indicating job satisfaction would be incorporated and tested.