Home

Menu

Loading wiki pages...

View
Wiki Version:
<p>In our analysis plan, we indicated the analyses mentioned below. We will provide a brief summary of the results so far. </p> <p><strong>1. We did both a MANOVA and ANOVA to compare groups on both the separated communality items (attractiveness and quality separated) and the communality_total scores.</strong></p> <ul> <li>Our condition variable significantly affected our communality_total scores; F(1, 113) = 12.137, p &lt; .001, p-eta sq = .18 (this effect did not remarkably change if we omitted those who are actively looking for a house (F(1, 113) = 11.19, p &lt; .001) or when controlling for sex (F(1, 113) = 11.45, p &lt; .001).</li> <li>A comparable effect emerged on our communality scores; F(1, 113) = 12.561, p &lt; .001, p-eta sq = .19</li> <li>On communality_total, the "outside" condition was higher than the "going inside" condition (t(113) = 3.63, p &lt; .001) and the control condition (t(113) = 4.70, p &lt; .001). The "going inside" and control conditions were not significantly different (t(113) = 1.07, p = .286). In a weighted contrast, the "outside condition" was also significantly higher than the "going inside" and control conditions together, F(1, 113) = 23.10, p &lt; .001, p-eta sq = .17.</li> </ul> <p>Note: we are not yet reporting on attractiveness and quality, because N is too small to conclude differences with communality_total (note: we also have a separate intention to purchase variable). </p> <p><strong>2. We also utilized an ANOVA to compare groups on need for affiliation.</strong></p> <ul> <li>Our condition variable also significantly affected our NFA scores; F(1, 113) = 3.67, p = .029, p-eta sq = .062 (this effect did not remarkably change if we omitted those who were actively looking for a house (F (1, 113) = 2.81, p = .065) or when controlling for sex (F(1, 113) = 3.73, p = .027). </li> <li>The "outside" condition however was not significantly different from the "going inside" condition (t(113) = 1.46, p = .147), but it was higher than the control condition (t(113) = 2.71, p = .008). There was no significant difference between the "going inside" condition and the control condition (t(113) = 1.24, p = .216). In a weighted contrast, the "outside condition" was also significantly higher than the "going inside" and control conditions together, F(1, 113) = 2.71, p = .008, p-eta sq = .062.</li> </ul> <p><strong>3. To discover if the proposed mediating effect between the relationship between temperature and the evaluation of the house via Need for Affiliation was present, we leveraged a bootstrap procedure using Preacher and Hayes' (2008) INDIRECT macro, with 1000 resamples.</strong></p> <p>The mediation analysis also revealed no effect of NFA onto communality_total (t(113) = 1.77, p = .08), and a CI 95 overlapping 0 (-.0008 .0908). Given the values of the effect, we will run these analyses again once we have collected the entire sample, and we expect to find the full mediation there. </p> <p><strong>4. Finally, using the PROCESS macro by Hayes (2013), we explored whether daily temperature played a moderating role in our proposed effects.</strong></p> <p>We ran Model 8 from Hayes' (2013) PROCESS Macro, and observed a significant interaction with outside temperature (t(113) = 2.67, p = .009). This interaction means that the observed mediation pans out at M_t (11.58), t(113) = 3.94, p &lt; .001 and M+SD_t (17.93), t(113) = 3.26, p = .002, but not at M-SD_t (5.22), t(113) = .11, p = .91. </p> <p>We suspect that this may be due to a floor effect (in that low temperatures elicit NFA irrespective of being inside) or simply low power. This will be investigated with the full sample. </p>
OSF does not support the use of Internet Explorer. For optimal performance, please switch to another browser.
Accept
This website relies on cookies to help provide a better user experience. By clicking Accept or continuing to use the site, you agree. For more information, see our Privacy Policy and information on cookie use.
Accept
×

Start managing your projects on the OSF today.

Free and easy to use, the Open Science Framework supports the entire research lifecycle: planning, execution, reporting, archiving, and discovery.