Revised Shyness Scale

Shyness relates to the ‘self-esteem, anxiety and fearfulness that an individual experiences when faced with social situations’ (Jones & Carpenter, 1986). As shyness is often conceived as an aspect of personality, psychologists often use personality as a mechanism to understand the different aspects of shyness. Personality can be defined as a ‘dynamic organisation inside the person, of psychological systems that create the person’s characteristic patterns of behaviour, thoughts and feelings’ (Allport, 1961). As a result of this relationship personality psychologists have examined the methods available to establish a link between these two dimensions in order to investigate how they affect our social behaviour and relationships.
Personality

The Likert Scale scores items using 5 ratings (1 = very uncharacteristic to 5 = very characteristic). The RCBS is used globally as it is found to have strong internal consistency (α = .86) and test-retest reliability (r = .88) (Hopko, Stowell, Jones, Armento, & Cheek, 2005).
The scale we developed is a 12 item shyness scale. It contains three subscales: (social, stranger and assertiveness difficulty). Participants were informed that the scale was created in order to examine what attitudes people hold and what emotions they experience. Beside each item, there was space provided for the participant to rate the item using the Likert scale (1 = strongly disagree to 5 = strongly agree). Participants were also asked to record their age and gender.
After the 190 participants (N = 190) completed the scale their data was combined to create one dataset. After this a reliability, validity and factor analysis was carried out on the data. A reliability analysis was utilized to investigate if the test efficiently assessed the construct being measured. A validity analysis allowed us to investigate if there was a relationship between the subscales of our scale and the other scales participants had to answer. Finally, a factor analysis established whether one or more of the items underlie other

To calculate the internal reliability of a scale Cronbachs alpha, α, is used. Cronbach’s alpha allows us to calculate the variance within an item and the covariance between any two items on the scale. For a scale to be deemed reliable α > 0.8. However, .7 is also acceptable.
After testing the reliability, the α for all 12 items of the scale was revealed to be .82 which indicates good reliability. To investigate the reliability of the scale further it was broken into 3 subscales; social, stranger and assertiveness.
In relation to the three subscales social items had an α of .5, which means that this subscale is not reliable. In the table ‘Item-total statistics’ there is a column labelled ‘corrected Item-Total correlation’ This contains the correlations between each item and the total score for the subscale. If any of the items have less than .3 of a value, it means that this item does not correlate very well with the scale overall. Item 7 “I feel uncomfortable in familiar group settings” has a score of .26 indicating that it is not stable. However, Item 10 “I look forward to attending social events with my family” has a score of .14, much lower than a required score of .3. Also, if item 10 was deleted the α would increase to .56 improving reliability. Therefore, this item should be deleted from the