Forming impressions of hotels: the subtle power of words

Abstract

The aim of the study was to investigate the potential power that subtle wording changes may have in altering the impressions that potential guests gain from brief descriptions, such as on the home pages of hotel websites or in online reviews.

The intention was to determine whether the findings of earlier psychological research on the formation of impressions of people could be replicated regarding impression formation of hotels. In the psychological experiments, students first read a brief biography about a guest lecturer, then later attended a discussion led by him. They then rated their impressions of the man on a number of dimensions. All of the respondents experienced exactly the same discussion and received exactly the same biography. The words were identical except for one word. In one version a sentence said people who knew this man described him as cold. In the other version he was described as warm.

The change of this one word alone had significant effects on people’s ratings on some dimensions and very little effect on others. For example, those who received the warm version rated the man as more generous than those who received the cold version.

The current study was designed to see if this effect would occur when hotel management students formed an impression of a hotel, especially when there was limited information available – such as on the home page of a hotel’s website.

Data were collected from 60 students from a hotel management college who were familiar with hotels either as guests, students on industry placement, or both. The students were in their early twenties and represented a number of different nationalities.

Information from the home pages of two New Zealand hotel websites was adapted and edited to form a description of a fictitious hotel. The brief description of the hotel was followed by a survey containing 15 items, each describing a dimension of the hotel. The students gave each item a score of between 1 and 6 (see Table 1).

Respondents received exactly the same descriptions of the hotel except for the change of one word. In one version a sentence said, “Guests who have stayed at the hotel generally describe it as having a cold ambience.” The same sentence in the other version said, “warm ambience.” Thirty students read the warm version and the other 30 read the cold version.

Table 1: Statistically significant dimensions of ‘warm’ versus ‘cold’ hotel
Hotel dimension (scored 1→6)	‘Warm’ mean	‘Cold’ mean	t-test result (p)
Popular → Unpopular	2.0	3.3	0.002
Busy, bustling → Not busy, inactive	2.5	3.6	0.011
Quiet → Noisy	2.6	3.5	0.019
Tidy, well-kept → Untidy, run-down	2.0	2.8	0.035
Environmentally friendly → Not environmentally friendly	2.4	3.1	0.040
Modest → Extravagant	3.9	3.1	0.042
Knowledgeable staff → Not knowledgeable staff	2.9	3.3	0.050
Note: mean=average score (between 1 and 6); p=probability

Statistical analysis of the group averages revealed seven statistically significant dimensions. Table 1 shows that the ‘warm’ group perceived the hotel as more likely to be more popular, busier and bustling, quieter, tidier and well kept, environmentally friendly, extravagant and with more knowledgeable staff than did the ‘cold’ group.

Having rated the warm ambience as more likely to be bustling it might be expected that it would be rated noisier. That was not the case. Revisiting some students and further questioning them revealed that a cold ambience was perceived as more likely to have harsh surfaces such as tiles and glass that would echo noise. A warm ambience was perceived to be more likely to be carpeted, curtained and with décor that absorbed sound.

The findings show that the warm-cold effect could be replicated regarding impression formation of hotels and may suggest that some dimensions are core for forming an impression and others are peripheral. It appears ambience described as warm and cold may be among the core dimensions. The current study was an initial exploration of this phenomenon and further research would be required to identify whether other dimensions are core or peripheral. However, a more recent replication of the study by the current author using “welcoming – unwelcoming” produced five statistically different dimensions and hence may also be a core dimension. Whatever the case, the findings suggest that care needs to be taken when hoteliers choose wording to include in descriptions on the home pages of their websites.

Corresponding author

Robin Hill can be contacted at: robinh@pihms.ac.nz