Popular in PSYC4201W
verified elite notetaker
Popular in Psychlogy
This 10 page Class Notes was uploaded by Rachel Onefater on Thursday April 7, 2016. The Class Notes belongs to 76884 at George Washington University taught by Dr. George Howe in Spring 2016. Since its upload, it has received 15 views. For similar materials see PSYC4201W in Psychlogy at George Washington University.
Reviews for Lecture Notes
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 04/07/16
Statistics for testing group comparability and effects of intervention Groups are essentially equivalent for anything that might confound the studyfor the apps. Are groups balanced on potential confounds? • Random assignment should reduce group differences on confounds • But that is probabilistic, not exact • Good to test whether groups are comparable on: – Pretest measures – Other potential confounds Problem of multiple comparisons • With two groups, we simply compare means with ttest → How do we get Standard Error, or if we are outside the bounds of the hypothesis(null hypothesis). • With more groups, we could do multiple pairwise comparisons (def.) Compare one app. To another app. In different orders and combinations – 3 groups: 3 comparisons – 4 groups: 6 comparisons – 5 groups: 10 comparisons • Problem: when doing multiple comparisons, the probability grows that any one comparison will be significant by chance – Probability levels set for significance (such as .05) are no longer correct Ex: Probability of the null hypothesis is less than .05. If you were to do two tests, even if you set hat same range, it would not be at the .05 level, it would be more likely than .05 because each time you do a test it increases the probability of finding something (i.e. if you test 50 things vs. 2 things) → This capitalizes on chance. Two solutions • Analysis of variance • Bonferroni corrections Variability among means • Consider a study randomly assigning to 5 intervention programs – We want to test whether these groups are equivalent on some potential confound, such as Type A behavior – We can calculate a mean for each group means can fluctuate, and do they differ enough that there is a significant difference between/among those means – There will be variability among those 5 means do to sampling error – How do we know whether this variability is greater than chance? Analysis of variance • Analysis of variance: a method of testing whether variability among means differs from zero • We can calculate a variance for the set of group means • We can calculate a standard error for that variance • We can use this to test whether the variance is far enough away from zero to conclude that it is not due to chance • Calculating standard error of variance among means is complex; depends on – Variance among the means – Variance within each group contributing a mean Ftest • Ftes: hypothesis that a set of means are equal – Assumes means are normally distributed – Assumes that variances within each group are equal across groups these are assumptions that one can test, and you can calculate a value. • F value: statistic calculated using analysis of variance • Null hypothesis: hypothesis that means are all equal • P value: probability of null hypothesisthe lower the p value, the lower the probability of the null hypothesis – The lower the p value, the less likely the null hypothesis; that is, the more likely that means differ *NOTE: similar to t value in how far the value is from 0, bt the calculation is a bit different. Analysis of variance in Excel • Setting data up – Put columns of data for each group next to each other → Want to see if the means on these three groups differ. Analysis of variance in excel(Stepbystep) • Click on Data tab, then Data Analysis icon • Select ANOVA (i.e. Analysis of Variance): Single Factor, and click OK • Click onto red arrow next to Input Range, and highlight all columns of data, including headings. Hit enter. • Make sure the Labels in First Row is checked • Select New Worksheet Ply, and type name in (ANOVA Type A) • Click OK, **Change to three decimal places • Summary of analysis of variance will appear in new worksheet tab ANOVA results • Ignore most of details • P value gives significance level – P < .05 indicates there are significant differences among groups **Very likely to have variability of means Reporting tests of equivalence • Common practice: report group means, standard deviations, F value, p value • (Remember, SD is square root of variance) “We ran this analysis of variance to see the effects of confounds on variables…” Strengths/Weaknesses of ANOVA • Strength – Tests whether there are differences, regardless of number of groups→ not capitalizing on chance by doing multiple tests! – Only one test regardless of number of groups, so we are not capitalizing on chance • Weakness – Cannot tell us which groups, specifically, differ from one another → This is something you may want to check on and/or interpret! Testing specific group differences • To test specific group differences, back to ttest • However, need to account for increased chance findings • One solution: Bonferroni correction (more conservative approach) – Determine number of comparisons, divide p value threshold by that number – This gives a new p value threshold for significance → Correct the threshold value for saying if something is significant or not. Divide .05 by the number of comparisons you are going to do! Combining ANOVA and paired comparisons • Strategy: – Conduct ANOVA – If F is not significant, interpret this as no significant differences among group means, and stop here – If F is significant, use ttests to compare means of interest, and use Bonferroni adjustment to adjust threshold p values for interpreting an effect as statistically significant – Remember: this adjustment is conservative: you can find significant F, but no significant pairwise comparisons! *NOTE: If we find it significant, we go further. Example: comparing means on outcomes • In your trial of stress management apps, we want to compare all three means: – App 1 vs app 2 – App 2 vs app 3 – App 1 vs app 3 • Back to ttests Let excel do the whole ttest (stepbystep) • In excel, click on Data tab, then Data Analysis icon • Select “ttest: Two sample assuming equal variance” , OK • As usual, enter variable one range(Type A Personality measure), variable two range (including labels), click on “labels”, and add a new worksheet ply Output: • Compare twotail p to adjusted threshold • Here, we’re doing three comparisons, so threshold is .0167 • P value (.002) is lower than .0167, so means are significantly different → So this far all the apps(the comparisons between the apps) *NOTE: It’s ok to have a negative tstatistic because this just means that the first one is smaller than the second one Reporting mean differences • Common approach: table of means with subscripts indicating which are different: → Group(App.) 2 was different from Group 1 and from Group 3pretty standard we are now finding that app 1 and app 3 had more positive affect and app. 2, not so much! Summary • For your report on study 3: • Conduct ANOVAs for four potential confounds, and summarize in table – Type A – Loneliness – Pretest of positive affect – Pretest of negative affect • Conduct ANOVAs for outcomes (posttest negative affect, posttest positive affect – Report means, F, p value in table – If ANOVA is significant, conduct ttests to compare all three pairs of means, and use subscripts in table to indicate which differ – If ANOVA is not significant, give all means the same subscript (they don’t differ) Mindfulness based stress reduction → Moved psychology back to the more experiential/phenomenology. What ways do we experience things. CBT is focused more on personal, lived experiences. Moving away from biological and away from the objective focus of specific observable behavior. What is going on here and how can I use that information? 1. How does our personal experience translate to other people 2. How can we transfer these experiences to scientific knowledge Origins • John KabatZinn→ University of Massachusetts medical school → Asian Culture/Stress Reduction and bringing that to Western Society. *Goal: Bring life philosophies that guided those meditation practices – Stress reduction clinic – Center for mindfulness in medicine • Early focus: patients with chronic illnesses (example: chronic pain, psoriasis, surgery, unknown etiology) → We still do not have a real clear understanding on the neurological substrate of pain. → People can be put in positions in which they can be in greater control of pain(i.e. hypnosis). → Pain is an experience, we cannot yet quantify it in a biologically meaningful waythe more powerful medication is, the more addictive it is. a. 8 week series, people came in once a week, but agreed to do things outside on a weekly basis i. Included a full one day retreatfairly intensive What does MBSR cover? • Alidina (2015). The mindful way through stress. • Principles of MBSR – Intentionality(def.) need to form an intention, selfmotivating – Paying attention (def.) a focus – In the present moment – With an attitude of acceptance, curiosity, self compassion, and openness (def.) not about fate, but to accept what come without being entrapped by that thoughtbeing fused with your thoughts – n meditation, and in everyday life MBSR practices • From Alidina (2015) Mindfulness boosters (bringing mindfulness to daily life) • Mindful showering – Set an intention to be mindful – Mix up your routine, so you remain more attentive – Attend in each moment to each of your senses, sound, sensation, vision • Mindful teeth brushing – Use your opposite hand, attend to each moment of each tooth • Mindful walking – Take a different route – Attend to each part of your step, each sensation Social mindfulness exercises *NOTE: Social World stresses us, and helps relieve us from stress • Mindful kindness – Plan a small act of kindness or helpfulness – Attend to your thoughts, feelings during that time with gentle curiosity • Mindful gratitude – Spend time thinking about something that you are grateful for – Attend to your thoughts, feelings • Mindful communication – Use mindful awareness when talking with someone else – Stop, look, listen, avoid interrupting – Keep an open body language and be curious → One of the signature symptoms of social anxiety, people are unlikely to watch or attend to others, particularly look them in the eye. The ABC approach to stress • Awareness – Notice your stress signs – Identify your stressors • Breath and beliefs – Feel your breath and body – Reframe beliefs about the stress or stressor • Choice – Accept situation or – Change situation Expanding Awareness Meditation Alidina (2015) Why does mindfulness work? • Goldin et al (2016) Group CBT Versus MBSR for Social Anxiety Disorder: A Randomized Controlled Trial. • Assigned patients with diagnosis of Social Anxiety Disorder to: – MBSR group – CBT group – Wait List control Targets of MBSR intervention • Specific mindfulness skills – Observing – Describing – Acting With Awareness – Nonreactivity to Inner Experience – Non Judging of Inner Experience • Attentional focusing • Attentional shifting • Rumination Targets of CBT intervention • Frequency of cognitive reappraisal (def.) How can they shift their way of thinking • Sense of selfefficacy about cognitive reappraisal • Frequency of subtle avoidance or safety behavior • Cognitive distortions Intervention effects on social anxiety • Both CBT and MBSR significantly reduced anxiety compared to waitlist, and both maintained effects • No differences between CBT and MBSR *NOTE: results suggest that these have efficacy MBSR effects on targets • MBSR significantly changed all targets except attention shifting MBSR: did targets mediate effect on anxiety? • MBSR showed mediation through six of eight targets CBT effects on targets • CBT significantly changed all targets CBT: did targets mediate effect on anxiety? • CBT showed mediation through all targets Differences? • MBSR and CBT equally effective at changing all targets except: – CBT slightly stronger on decreasing subtle avoidance behavior – Mediating effects of reappraisal efficacy and subtle avoidance a bit stronger for CBT Conclusions • Similar efficacy (immediate and at 12 month followup), similar impact on targets, subtle differences in mediation • Possible that different people prefer different approaches (although no differential dropout) • We may need better measures of target processes to tease apart different effects – Participants in tango class showed changes comparable to mindfulness intervention (Pinniger et al., 2012) – Measures may be vulnerable to social desirability: binge drinkers’ scores higher than those of people at mindfulness retreat (Leigh et al, 2005) *Awareness componentspeople are asked to be aware of what is going on in their lives to note for any changes MBSR for health care providers? • Can MBSR reduce risk of burnout for providers who often face highstress contexts? • Lamothe (2016) Outcomes of MBSR or MBSRbased interventions in health care providers: A systematic review with a focus on empathy and emotional competencies. • Reviewed efficacy for 39 trials with – Medical students – Psychiatric inpatient staff – Nurses – Physicians • Positive effects on – Reducing burnout – Decreasing anxiety, depression – Increasing empathy • No clear evidence on other emotional competencies *NOTE: Stress relief, even for people who treat stress is very important! Summary • Substantial evidence for efficacy of MBSR • Differential efficacy (value added) still unclear • Work on mediators (why it works) and moderators (for whom it works best) still in early stages → everyone’s individual experiences with mindfulness intervention is different for different people and may depend on stressors, personality etc.
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'