Perceived support: meaning and application

You ask people to do something new, and whether they go along also depends on whether they expect to get help. That estimate is called perceived support. You will read where the concept comes from, how it differs from related concepts, how to research it and where its limits lie.

Perceived support is the extent to which people expect to receive support from the organisation, their manager, colleagues or their wider environment when they make a choice or face a change.

It concerns an expectation and an experience, not the offer itself. An organisation can arrange a lot of help while employees do not see it or do not count on it. The concept builds on research into perceived organisational support and perceived social support.

The concept has two roots. The first is in work and organisational psychology. In 1986, Robert Eisenberger and colleagues introduced perceived organizational support (POS): employees' general belief about the extent to which the organisation values their contribution and cares about their well-being (Eisenberger et al., 1986). This organizational support theory assumes that employees who experience support feel obliged to give something back. A review of more than seventy studies found that fair treatment, supervisor support and favourable rewards and job conditions were associated with POS, and POS in turn with outcomes such as commitment, performance and lower withdrawal (Rhoades & Eisenberger, 2002). A later meta-analysis of 558 studies supported the theory's predictions about the antecedents and consequences of POS (Kurtessis et al., 2017).

A related concept is perceived supervisor support: support experienced from the direct manager. In the first of three studies, it was related to change in POS over time; in the third, its link with lower turnover ran entirely through POS (Eisenberger et al., 2002).

The second root lies in research on social support and stress. In 1985, Sheldon Cohen and Thomas Wills described the buffering hypothesis: support protects people mainly when they are under pressure (Cohen & Wills, 1985). They found evidence for such a buffering effect mainly when measures assessed the perceived availability of support that matches what the stressful situation demands.

To make the concept usable, we distinguish three layers. This division is an editorial working framework of Neurofactor; the last two layers follow the scientific literature.

• Available support is what has actually been arranged: a contact person, a training session, a helpdesk, extra time.
• Received support is the help someone has actually received during a certain period.
• Perceived support is the expectation that help will be there when you need it.

These layers do not coincide. A meta-analysis of 23 studies found an average correlation of r = 0.35 between self-reported received and perceived support (Haber et al., 2007). Bert Uchino argues for always measuring the two separately, because they can have different antecedents and mechanisms (Uchino, 2009).

Received support is not automatically beneficial either. In a diary study of couples in which one partner was preparing for a demanding exam, support that the recipient did not notice turned out to be the most beneficial for adjustment to stress (Bolger et al., 2000).

Forms of support

Cohen and Wills distinguish, among others, three forms that recur in much research:

• Instrumental support: practical help, resources or services, such as someone who watches over the first steps.
• Informational support: help in understanding and coping with a problem, such as explanation or advice.
• Emotional support: the signal that someone is valued and accepted; Cohen and Wills call this esteem support.

They also mention social companionship. Which form counts depends on the situation. Someone learning to use a new device may mainly need instrumental and informational support; someone who doubts their own ability may also need emotional support.

What perceived support is not

Perceived support differs from perceived social norms, which concern what others do or expect, not the help you expect to receive. It is also not experienced autonomy (how much control you experience) and not a motivational profile (which motives weigh most). People who expect to be on their own may experience a change as a threat sooner; see also desired state and threat state.

The job demands-resources model (JD-R) by Arnold Bakker and Evangelia Demerouti provides a useful framework (Bakker & Demerouti, 2007). The model distinguishes job demands, which cost energy, from resources, which help achieve goals, reduce demands or stimulate growth. Support from supervisors and co-workers counts as a resource at the interpersonal level (Demerouti & Bakker, 2011).

A change often brings temporary extra demands: learning something new, correcting mistakes, losing pace. Within the JD-R framework, you can see perceived support as the expectation that resources will be available to meet those demands. That is an interpretation, not a proven mechanism for every choice. It does allow a testable question: does willingness to start change when people know what help is available?

Perceived support is an experience and is therefore usually measured through self-report. In organisations, POS is usually measured with questionnaires that ask about the organisation's appreciation and care. For perceived social support outside the work context, instruments include the Multidimensional Scale of Perceived Social Support (Zimet et al., 1988).

For a specific choice or change, write statements about the specific situation and distinguish by source (organisation, manager, colleagues) and form (instrumental, informational, emotional). An adapted or new questionnaire is not automatically validated: check reliability and structure in your own sample.

Also map the available support and, where relevant, ask about received support. Only then can you tell whether help is missing or simply not noticed. Open questions and free associations show which words people themselves connect with the change, such as "alone", "helpdesk" or "no time". See How do you measure associations?.

1. Describe the choice or change. What extra demands does it bring, and when?
2. Take stock of available support. Who helps, with what, at which moment?
3. Measure perceived and received support separately. Distinguish source and form; see also investigating the cause of the main pain point.
4. Only make real support visible. Name a specific contact person and help during the start-up, but promise nothing that is not there.
5. Compare variants. Test messages with and without an explanation of support on outcomes defined in advance.
6. Record the evidence. Note source, time of measurement and uncertainty; see what evidence means in a target group profile.

This example is fictional and intended as an illustration.

Situation. A logistics company asks warehouse staff to work with a new handheld scanner. During the first weeks, this costs pace.

Choice question. How can the company increase willingness to use the scanner immediately and properly, instead of falling back on the old way of working?

Available information. On each shift, one colleague has been trained as a fixed contact person, and there is a helpdesk. Conversations show that many employees do not know the contact person and fear that their manager will keep expecting the old pace during the start-up.

Suitable approach. The company measures perceived support by source and form in a sample, using pretested statements, and takes stock of the available help. It then compares two announcements: a general text and a text that names the contact person on each shift, explains what help is available in the first week and what managers expect during that period. That last promise is only included if the company actually keeps it. The company defines in advance which outcomes count.

Possible interpretation. If employees perceive more support with the second text, that suggests unfamiliarity with existing help played a role. It is not proof that the scanner will therefore be adopted faster.

Next step. Compare usage data from both groups in a limited trial. No outcomes are known in this example.

• Equating the offer with the experience. A helpdesk or training does not guarantee that people expect support.
• Using perceived and received support interchangeably. They are moderately related, and each needs its own measurement.
• Reading self-report as objective fact. Answers reflect how people see the situation and what they are willing to report.
• Forgetting reciprocity. According to the theory, perceived support creates a sense of obligation. The relationship between support and commitment can therefore run both ways, and a score also reflects the history of the relationship.
• Reading correlation as cause. Much research is correlational. A causal claim requires an experimental comparison or at least repeated measurements.
• Assuming more visible help is always better. Visible support can also have costs; show clearly that help is available without imposing help on people.

Perceived support describes whether people expect not to be on their own when facing a choice or change. The concept is useful when you separate it from the support that actually exists and the support someone has already received. Research on organisational and social support shows consistent associations, but mostly based on self-report and correlations. So take stock of available support, measure the experience within your own target group and test whether making support visible actually makes a difference.