Understanding the Housing Aspirations of People in Scotland

Appendix

Research Methods

This study adopted a qualitative methodology, involving interview and focus group research with 80 people across Scotland between February - June 2015. The sample of participants was recruited to reflect Scotland's diverse geographies and socio-economic variances in Scotland's citizens. As outlined in Chapter 1 the aim of this project is to provide in-depth information to help understand the housing aspirations of people in Scotland now and in the future. Whilst Communities Analytical Services has a wealth of quantitative data (from survey based research) on housing aspirations, this study provides a rich and nuanced qualitative perspective, thus contextualising people's aspirations within the wider social, economic, political and cultural context of the nation.

Participants were offered a choice of how they wished to get involved in our research:

• Semi-structured interviews: either by telephone (n=64), face-to-face (n=4) or via SKYPE (n=0)
• Focus groups: either face-to-face (2 FGs, n=10) or online in an asynchronous focus group (1 FG, n=2), using the dedicated Visionslive online web-interface.^[9]

The rationale for having flexibility in our research methods was to maximise recruitment and avoid any potential digital exclusion that might arise from relying on online methods alone. Despite our past experience in using online focus groups in housing research with young people (Moore et al 2015), there was little interest in this method amongst our participants in this project. This may reflect the wider age range of the sample (30 per cent were over 65), as well as the tighter fieldwork timeline involved.

The project was informed by purposive sampling, which is a common approach in qualitative research. It requires selecting participants relevant to the criteria we are interested in, and which enable us to answer our research questions. The aim here is not to create a representative sample, for qualitative research is not guided by the principles of statistical inference that is common to quantitative studies. Rather it is about ensuring a sufficient diversity in the sample to address our research questions by ensuring we can explore key characteristics we are interested in. Moreover, sampling frames are only ever a guide in qualitative research: a target to aim for; ultimately sample size is linked to 'saturation point' - data collection ends when new data fails to shed further light on the issue under investigation.

To this end, we adopted a two-pronged approach to participant recruitment that was sensitive to the need to recruit different types of people from different geographical locations. This operated at two scales: the national (Scotland) level and local authority (case study) level:

• Nationally: we contacted past respondents to the Scottish House Condition Survey who have given permission to be re-contacted for future research (n=500). The sampling frame was drawn from across Scotland and designed to target those individuals with the characteristics we were interested in. We received a response rate of 6 per cent (n=30). The majority of these respondents were in the older age range (over 65) and were outright homeowners. To compensate for this we advertised for further participants online (website, Twitter) and through our network of gatekeeper organisations (e.g. social landlords, advice centres). We received a further 15 participants through this route from across Scotland, who were drawn from both rural and urban locations.
• Local Authority Case Studies: we used a dedicated recruitment agency to recruit 35 people from across 5 local authority case studies in Aberdeen City, Argyll & Bute, Perth & Kinross, Renfrewshire, and the Scottish Borders. These locations were chosen to reflect variances across Scotland in terms of house prices and housing tenure structure, as well as to allow for an exploration of the differences between rural and urban areas. More details on the case studies can be found in the next sub-section.

Table 2 Sample by Case Study Location

In addition to geography, our sampling matrix was designed to reflect socio-economic variances across Scotland and was diversified by housing tenure, age, gender, ethnicity and also included individuals who self-identified as having a long-term illness or disability (see Table 3). A number of demographic groupings were identified at the outset of the project as of being of interest to policy:

• Older people
• Young people
• The squeezed middle
• People with disabilities
• People living in rural areas
• People living in (potential) regeneration areas

Our sampling strategy successfully recruited participants from each of these groups with the exception of those living in (potential) regeneration areas. Such individuals are difficult to recruit unless specific geographical locations are identified at the outset drawing on knowledge and understanding of the micro-geographies of local housing markets.

Our research methods produced verbatim transcripts, which were imported and analysed in the Computer-Aided Qualitative Analysis Software (CAQDAS) NVivo. Guided by the principles of Grounded Theory (Charmaz 2006), analysis and fieldwork were conducted in tandem, allowing for emerging themes and concepts to be identified and refined through the project's lifespan.

Table 3 Sample Characteristics

Local Authority Case Studies

Participants were sampled from 5 local authority case studies, which included a mix of:

• rural and urban areas
• housing tenure structures
• local authorities with different patterns of income inequalities
• local authorities with different age profiles

These case studies included:

• Aberdeen City: is an urban local authority in the north-east of Scotland, and Scotland's third largest city. House prices in Aberdeen are amongst the highest in Britain, not just Scotland. The city has a strong economy, fuelled by the oil and gas industry, and more recently the renewables sector. This influx of economic migrants to the city (from within and beyond the UK) has further contributed to a thriving private rented sector, with rents at the top end being amongst the most expensive in Scotland. The city also has a vibrant student population. Nonetheless homeownership remains the largest tenure, although slightly below the national average.
• Argyll and Bute: this rural local authority in the west of Scotland has a geographically dispersed population in urban, rural and island communities. Two-thirds of households are in the private homeownership sector, with a significant percentage being owned outright. However the local authority also has one of the highest levels of ineffective housing: that is housing which is unavailable to local people because it is vacant, a second home or a holiday let. This makes it difficult for local people to remain in the area. In addition, across all tenures poor condition is an issue with nearly 90 percent of housing failing to meet the Scottish Housing Quality Standard. The issue of poor condition is complicated further by an increasingly ageing population.
• Renfrewshire: is an urban local authority in the west of Scotland that shares a boundary with Glasgow. Levels of social housing and homeownership are higher than the national average. Although the size of the private rented sector is smaller than across Scotland as whole, it has nonetheless doubled in the last decade, mirroring national trends. Average house price sales are lower than the Scottish average, and levels of sales have dropped considerably over the last decade. Renfrewshire's overall population change has experienced a small decline since 2001, and levels of deprivation are shown to be above the Scottish average.
• Scottish Borders: a rural local authority that shares borders with a number of Scottish council areas, and also with the English counties of Cumbria and Northumberland. Although unemployment is relatively low, wages are amongst the lowest in the country, which in turn has implications for the type of housing people can afford. Homeownership is expensive due to a decline in new housing supply, and commuters being attracted to lower prices relative to Edinburgh and the Lothians. The private rented sector has also grown significantly over the last decade and is now slightly above the national average. Over 70 percent of the housing stock fails the Scottish Housing Quality Standard, with empty homes also an issue in the locality.
• Perth & Kinross: around two-thirds of the population lives within the city of Perth or its surrounding commuter villages. There are however also sizeable settlements in the more rural parts of this local authority, which is situated in central Scotland. Levels of homeownership are higher, and levels of social renting are lower than the national average. In addition levels of outright homeownership are high, as is the number of second and uninhabited homes. Affordability is a significant issue across the local authority area, and average house price sales are well above the national average, which creates challenges for first-time-buyers in particular. As a result the PRS plays a significant role in housing local people. In terms of deprivation indices, figures for Perth & Kinross are below the national average.

How to access background or source data

The data collected for this social research publication may be made available on request, subject to consideration of legal and ethical factors. Please contact Julie.Guy@gov.scot for further information.