PART 1: TESTING HYPOTHESES OF DIFFERENCES IN MEANS
Part 1
Data file for analysis: The data file is called Differences assignment.sav and is available in the assignment folder on BB.The data was generated by the tutor and relates to a fictitious experimental study described below.
Research study description
The study focused on examining the relationship between reward sensitivity and food-cue exposure and how they could independently or co-jointly explain food cue reactivity, response inhibition and food intake.
Independent variables: Sixty participants were recruited and tested for how sensitive they are to reward and subsequently divided into those who were high (n=30) or low (n=30) in reward sensitivity. This variable is labelled "Group" in the data file. Each group was then randomly allocated to an experimental condition where participants were EITHER exposed to very tasty and desirable foods (chocolate, cakes) OR very bland foods (stale bread, cold rice). This variable is labelled "Exposure" and has two levels: 1) Bland cue and 2) Tasty cue. The breakdown of the sample is illustrated in the SPSS cross tabulation below. All participants were assessed before and after exposure to both types of cues, although this variable is a within-subjects variable and so appears as pre- and post-measures (see dependent variables)
Dependent variables
- Appetitive stateand reaction: assessed before and after exposure to both types of cues in terms of how hungry they felt (100 = very hungry, 0 = not at all hungry) and how strong their desire to eat was (100 = very strong desire to eat, 0 = no desire to eat). These variables are labelled: pre-hungry and post-hungry, pre-desire and post-desire.
- Response inhibition: Following the exposure to food cues, and after making post-cue ratings, all participants took part in a response-inhibition task (Stop-signal task) which generates a Reaction Time in milliseconds. The longer the reaction time, the weaker the response inhibition. This is labelled SSRT (msec) in the data file.
- Food intake: after completing the Stop-signal task participants were offered opportunity to eat some tasty food (chocolate-chip cookies) while waiting for the researcher to complete some admin. This was a cover story as we wanted to see how much they might eat. The amount eaten (in grams) is labelled Intake (g) in the data file.
Covariate
- Participants were assessed prior to arriving at laboratory in terms of self-reported impulsivity which is known to be related to behavioural measures of response inhibition. This is labelled Impulsivity in the data file
Hypotheses
Hypothesis 1: Following exposure to tasty food cues,ratings of appetitive reactionswill be stronger than afterexposure to bland food cues, and this will be more pronounced for the high-reward sensitive group than the low-reward sensitive group (hint: is there repeated measurement?)
Hypothesis 2: Exposure to tasty food cues will result in a weaker inhibitory control than exposure to bland food cues and this will be more pronounced for the high-reward sensitive group than the low-reward sensitive group (additionally one would want to examine whether controlling for baseline impulsivity improved estimation of F-ratios)
Hypothesis 3: Exposure to tasty food cues will result in greater food intake than exposure to bland food cues and this will be more pronounced for the high-reward sensitive group than the low-reward sensitive group
Decide on the most appropriate analysis method for each hypothesis and then provide short answers to each question using numerical and graphical evidence where appropriate to support your answer. Number your answers clearly
Task
1 Check statistical assumptions that need to be met for hunger and desire variables.
Describe the outcome of your analysis of assumptions and evidence conclusions reached by descriptive interpretation of graphical and numerical information
2 Check statistical assumptions that need to be met for SSRT variable.
Describe the outcome of your analysis of assumptions and evidence conclusions reached by descriptive interpretation of graphical and numerical information
3 Check statistical assumptions that need to be met for Intake variable.
Describe the outcome of your analysis of assumptions and evidence conclusions reached by descriptive interpretation of graphical and numerical information
4 Test hypothesis 1 - Appetitive reactions
Report the outcome of the testing in a descriptive narrative using as much numerical information as necessary to provide a complete picture of the outcome. Provide brief evaluative comment about support for hypothesis
5 Test hypothesis 2 - Response inhibition
Report the outcome of the testing in a descriptive narrative using as much numerical information as necessary to provide a complete picture of the outcome. Provide brief evaluative comment about support for hypothesis
6 Test hypothesis 3 - Food intake
Report the outcome of the testing in a descriptive narrative using as much numerical information as necessary to provide a complete picture of the outcome. Provide brief evaluative comment about support for hypothesis
7 Explain why a multivariate analysis cannot be conducted to assess SSRT and Intakeas dependent variables at the same time. Support your answer with evidence from analysis, and explanation derived from recommended sources
8 Re-test hypothesis 2 controlling for Impulsivity and describe using relevant values what the differences are when Impulsivity is used as a covariate. Demonstrate by evidence whether assumptions of linearity and homogeneity of regression are met
PART 2: TESTING HYPOTHESES OF ASSOCIATION AND PREDICTION
Data file for analysis: The data file is called Associations assignment.sav and is available in the assignment folder on BB. The data is comprised of merged files from a number of studies and was generated by the tutor.
Research study description
Mindfulness based therapies (MBTs) for eating disorders show potential benefit for outcomes yet evidence is scarce regarding the mechanisms by which they influence remission from symptoms. Maladaptive emotion regulation is a key psychological feature of all eating disorders. One way that mindfulness approaches create positive outcomes is through enhancement of emotion regulation skills, specifically those related to identification of internal emotional states and acceptance of difficult emotional states. The aim of the current study was to identify facets of emotion regulation involved in the relationship between mindfulness and maladaptive eating behaviours. A cross-sectional design was used to examine facets of emotion regulation through which dispositional mindfulness may operate to influence symptoms in university students who are recognised to be a population at greater risk of developing eating disorders than the general population. Female participants (N = 256; 18-24yrs) completed questionnaires that assess risk for developing eating disorders symptoms (anorexia and bulimia), mindfulness (Five Facet Mindfulness Questionnaire), and two aspects of emotion regulation skill: emotional clarity and non-acceptance of emotional states (Difficulties in Emotion Regulation Scale).
Variables
Mindfulness- scored on a likert type scale and total score produced, called Mind in data file. Higher scores = greater dispositional mindfulness
Eating disorder risk (EDR) - scored on a likert type scale and total score produced, called EatRisk in data file. Higher scores = greater risk for developing eating disorder symptoms
Emotion regulation: again, scored on likert scale but higher scores indicate greater emotion regulation difficulties. Two aspects measured:
1) emotional clarity - ability to recognize distinct emotions - higher scores indicate greater difficulty doing this (ERSec in data file);
2) non-acceptance of emotional states - higher scores indicate difficulty tolerating disturbing emotions (ERSna in data file)
Researcher's predictions
1. Greater mindfulness associated with lower EDR and better emotion regulation skills; lower EDR associated with better emotion regulation skills
2. Having controlled for association between mindfulness and EDR, then emotion regulation skills (ERSec AND ERSna) would still be associated with EDR (hint: hierarchical regression)
3. The relationship between mindfulness and EDR would be mediated by emotion regulation skills but no prediction is made about which one might be more important than the other (i.e., ERSec vs. ERSna)
- Hint: Use PROCESS MODEL 4
Provide short answers to each question using numerical and graphical evidence where appropriate to support your answer. Number your answers clearly
Q Task
1 Check statistical assumptions that need to be met for use of variables in correlation analysis.
Describe the outcome of your analysis of assumptions and evidence conclusions reached by descriptive interpretation of graphical and numerical information
2 Test prediction 1: Create a correlation matrix of association between variables and interpret the outcome in a short descriptive narrative using as much numerical information as necessary to provide a complete picture of the outcome. You can use a table and appropriate graphs to support your answer
3 Test Prediction 2: Run a Hierarchical regression with Mindfulness entered at Step 1 and both Emotion regulation variables entered at step 2
Report the outcome of the testing in a descriptive narrative using as much numerical information as necessary to provide a complete picture of the outcome regarding the Model summary and Parameter statistics. You can use a table to capture the model and parameter summary statistics. Provide brief evaluative comment about support or not for prediction 2
4-9 Checking regression assumptions: provide brief comment to indicate whether the assumptions are met or not and where appropriate provide numerical and/or graphical evidence to support your conclusion
4 Check of Regression Assumption 3: You should have independence of observations (i.e., independence of residuals)
5 Check of Regression Assumption 4: There needs to be a linear relationship between (a) the dependent variable and each of your independent variables, and (b) the dependent variable and the independent variables collectively
6 Check of Regression Assumption 5:Your data needs to show homoscedasticity of residuals (equal error variances)
7 Check of Regression Assumption 6: Your data must not show multicollinearity
8 Check of Regression Assumption 7: There should be no significant outliers, high leverage points or highly influential points
9 Check of Regression Assumption 8: You need to check that the residuals (errors) are approximately normally distributed
10 Test prediction 3: Run mediation analysis using PROCESS 4 model add-in.
Report outcome in terms of boot-strapped confidence intervals for direct effect, total indirect effect, separate indirect effects and contrast between indirect effects. Provide a brief evaluative comment to indicate what the mediation effects means if one is observed.
Attachment:- Quantitative assignment.zip