# 評估樣本數 ### 決定樣本數可行方式 1. collecting data from (an) almost the entire population 2. choosing a sample size based on resource constraints 3. performing an **a-priori power analysis** 4. planning for a desired accuracy 5. using heuristics 6. explicitly acknowledging the absence of a justification. ### 設定效果量的可行方式 1) what the smallest effect size of interest is 2) which minimal effect size will be statistically significant 3) which effect sizes they expect (and what they base these expectations on) 4) which effect sizes would be rejected based on a confidence interval around the effect size 5) which ranges of effects a study has sufficient power to detect based on a sensitivity power analysis 6) which effect sizes are plausible in a specific research area. - 參考文獻 > Lakens, D. (2022). Sample Size Justification. _Collabra: Psychology_, 8(1), 33267. https://doi.org/10.1525/collabra.33267 ## 效果量的計算公式 - 連續變項之間的相關係數:[Pearson's r](https://en.wikipedia.org/wiki/Pearson_correlation_coefficient) - 平均值差異:[Cohen's d](https://rpsychologist.com/cohend/) ## 實作目標 **關鍵** 假設條件決定效果量的設定方式, 顯著水準, 應達到的考驗力 ### Level 1 轉換研究資料為效果量 (1)從目標論文及公開資料,找出研究結果裡,計算效果量的必要元素。(2)解析研究者如何設定檢驗假設的條件。 1. 確認效果量類別及計算公式 2. 了解公式中的統計量數:平均數?標準誤? 3. 分解用來計算統計量數的研究資料 #### 目標論文的效果量報告資訊 - Hughes et al. (2021) [五大人格與社經狀態的關聯性](https://psyarxiv.com/4jema/) see Analysis plan: H1 ~ H3 - Kathawalla & Syed (2021) [美國穆斯林民眾的生活壓力與心理健康](https://osf.io/6efyw) see Data Analysis Plan - Li et al. (2022) [影響紀錄報導筆法的口語因素](https://osf.io/7qmje) see Method: Prior Power analysis - McCarthy et al. (2021) [敵意促發效應](https://psyarxiv.com/gxp7u/) see proposal “Sample Size Determination” <!--- - Brown and Strand (2019) McGurk Effect的自動化機制 ~原論文實驗1 & 2結果:"Congruent versus McGurk fusion analysis" & "McGurk fusion versus McGurk non-fusion analysis" - Goldberg and Carmichael (2017) 語言複雜效應 ~ 原論文"Tests of main hypotheses" - Przybylski and Weinstein (2019) 青少年接觸暴力電玩時間與攻擊行為的相關性 ~ 原論文Table 3 ---> ### Level 2 估計樣本數 從原論文預先註冊資訊,找出研究者決定樣本數的方法及理由。 1. 決定預期效果量~參考[如何評估及解讀效果量](https://rstat-project.github.io/seed_courses/Lecture01.html) 2. 指定需要的考驗力~參考[設計有高考驗力的研究](https://rstat-project.github.io/seed_courses/Lecture02.html) 3. 使用合適的方法估計樣本數 ## 可用資源 - [G*power](https://www.psychologie.hhu.de/arbeitsgruppen/allgemeine-psychologie-und-arbeitspsychologie/gpower) - [faux](https://debruine.github.io/faux/) ###### tags: `EXPPSY_Book`