Sample Psychology Paper on Individual Differences and Psychometrics

Overview

Conducting regression analysis is a very important part of research methods and often forms key analysis for dissertations and wider general research.   In broad terms, regression is a linear approach to modelling the relationship between a dependent variable  and one or more explanatory variables (or independent variables). The case of one explanatory variable is called simple linear regression. For more than one explanatory variable, the process is called multiple linear regression.

 

For an example of multiple linear regression, medical experts have recently been researching the increase in Type 2 Diabetes (the dependent variable) amongst the general population.  During research, medical researchers would want to know whether, say, body weight, stress, exercise, family history, high blood pressure and particular diets are risk factors or predictors of diabetes.  To this end, questionnaires are widely used in Psychology to gather such data amongst the population who may be showing symptoms of prediabetes and actual Type 2 diabetes.

 

Regression is an extension to a correlational design as it describes how an independent variable or several independent variables are numerically related to the dependent variable and will express the weight of each independent variable and thus to what extent they predict the dependent variable.

 

Regression has been commonly used for many applied areas of psychology.  For example, within organizational psychology leaders want to understand what variables predict job performance and indeed links have been found with work-life balance and working hours, resources at work, job-related stress, relationships with leaders/peers, and organisational change.  Similarly, in education experts regularly research the variables that could predict successful learning outcomes. Variables could include intrinsic motivation, engagement with the programme, design of the programme, relationship with tutors, general learning and social resources, and so on.

 

The focus of this regression exercise is to examine independent variables that could predict satisfaction at University and give you hands-on experience of running a regression from scratch.  The independent variables will be formed from the literature around individual differences – personality, cognition, engagement, and so on.

 

A separate lecture is scheduled to examine the theoretical links between them and University satisfaction. It is important that you attend as this informs the introduction, hypothesis and discussion sections of the written report.

 

Please remain in your allocated workshop for the practicals.

 

Your task

 

Your task is to firstly complete a series of questionnaires on Qualtrics.  This will be conducted during the first workshop for this practical.

 

A raw data file will be made available for you.  You will firstly check and clean the data looking for any errors. We will then work on reverse scoring any items, checking reliability of the questionnaires and compute total scores.   As regression is sensitive to any ‘extreme’ scores (outliers) a series of analyses will be run.  A correlation matrix will then be used to check for further assumptions and requirements for a regression analysis.

 

Important note:

Basic Analysis – All students must complete this for the assignment – conduct a standard ‘enter’ multiple regression: Satisfaction is the dependent variable, the five personality traits together with Need for Cognition and Engagement are the possible predictors.

 

Based on research by Judge et al (2002) you would hypothesize that extraversion, neuroticism, conscientiousness, and to a lesser extent agreeableness are predictors of satisfaction with university, extrapolating from what has been found with job satisfaction. In addition, associated readings confirm a rationale for inclusion of Need for Cognition (this can be understood as a dispositional variable i.e. relatively stable like personality and is linked to the openness dimension but more specific to academic study) and emotional intelligence.   Student engagement as measured in this study comprises cognitive and psychological dimensions.  Student engagement has been historically linked to satisfaction.

Hierarchical Regression

With satisfaction remaining as the dependent variable, any personality predictors together with engagement and emotional intelligence that satisfy the assumptions for regression (i.ie. multicollinearity, avoidance of singularity) are entered in the regression equation on step 1 and then step accordingly.  The order of steps or blocks should be pre-determined by underpinning theory informing the rationale for this.   There are different ways in which you could conceptualise how these variables relate to satisfaction but this would be informed by prior research and theory.

 

Tasks to be completed by groups on a week by week basis (approximately)

 

Please note:  The weeks are approximate and is dependent on individual pace of working. Individual guidance will be provided on individual progress.

Week 3-4 Analysis of data and checking assumptions for regression. Running a standard ‘enter’ regression
Week 7 Working with hierarchical regression
Week 8-9 Starting write up with a focus on results and methods sections
Week  10 Continuing work on writing up
Week 11 Drop in support workshops

 

 

Assessment

The assessment for this part of IDP is based on an individual component (50% of the overall module grade).  You can seek and receive feedback from all workshop tutors on progress during the workshops, you can also use the drop-in hours of Becky and Zara together with additional appointments for them nearer the time of submission.