3006PSY Analysis Of Psychometrically Sound Measure Assignemnt Sample

Introduction: 3006PSY Analysis Of Psychometrically Sound Measure

Access Free Samples Prepared by our Subject Matter Experts, known for offering the Best Online Assignment Help Services in Australia.

An essential task in psychological research that aims to evaluate the validity and reliability of a measurement tool is the examination of a psychometrically valid measure. The study has reliable techniques to measure and assess psychological constructs thanks to psychometrically good measurements. This analysis involves looking at important psychometric features including validity, which verifies the statistic's accuracy in reflecting the target construct, and reliability, which guarantees consistent results. Examining elements such as component framework, item discrimination, and answer selectivity also have a role in the overall assessment of a measure's quality. A thorough examination of psychometric qualities helps to improve the accuracy and reliability of psychological assessments by ensuring the validity of study findings and informing the creation and improvement of assessment tools. This procedure is essential for promoting trust in the reliability of science and the applicability of research findings in an area of psychology.

Overview

This study has been performed through the four stages of analysis which gives a thorough description of the psychometrically. The screening of data and cleaning are essential in the first phase of the psychometric assessment procedure to maintain the accuracy of the measuring tool. Reverse coding pertinent items is the first step, followed by getting the descriptive statistics and frequency for all variables and performing an initial examination to identify any missing or anomalous data.

The second phase is concerned with removing things that perform poorly using item examination and reliability testing (El-Den, et al. 2020). Items with uneven distributions, test bias, high correlations between items, and low correlations between individual items are all carefully eliminated. After that, In Stage 3, discriminating indicators for the scale are given using the method of extreme groups and corrected item-total correlations. The Standard Mistakes of Estimations (SEM) and certainty stretch for individual scores are determined in the last stage,

Analysis

Stage 1: screening of data and cleaning process

Reverse code

There is a need to recognize items that were negatively keyed in reverse coding. In the next process, it should be opened by SPSS and the dataset. Items with reverse codes are selected for transformation and then re-coded into Different Variables (Alita, Putra, & Darwis, 2021). Here it needs to be picked. The items they would like to decode in reverse can provide fresh identities for the variables that have been recorded.

Obtain descriptive analysis/frequencies

The careful evaluation of data that is missing, either through combining or averaging, is required for constructing an overall spectrum value for the main construct. Then the values obtain the following descriptive statistics and frequency information (Patil, & Modi, 2019). By clicking on "Analyze," "Descriptive Statistics," and then "Frequencies." it also can be chosen as every variable (item) examined in the study.

Initial inspection of data

To proceed with the analysis, the process can proceed by clicking "OK". Initial Data Review should be done to check the Lacking Data. It should check the results for any unusual or missing patterns. If necessary, deal with any unusual events or missing data. In the next step, creating an age variable based on the year of birth can be Checked and prepared by demographic items (Nasution, Harahap, & Ritonga, 2022). The steps to do this are ‘Transform to 'Compute Variable'.

Prepare demographic items

Figure 1: Creating of demographic items

(Source: Self-created in SPSS)

In the next step, the original values should be listed in the "Old Values" column and the reversed codes in the "New Value'' column. Then it should have to Press "OK" to put the changes into effect (Purwanto, 2021). Age and gender demographic characteristics are transformed and dummy-coded in preparation for future analysis.

Total scale-score

Figure 2: Total score calculation

(Source: Self-Created in SPSS)

To Recode into Different Variables it is needed to make a new parameter for gender that is dummy-coded (female = one, male/other = 0). To create the following total scale score, if there is no missing data, then it should be chosen “Transform" and then "Compute Variable." SUM (q1, q2,...) is the formula to use. If any data is missing, use the 'Transform' -> 'Compute Variable' command (Ntumi, Agbenyo, & Bulala, 2023). Then the command should be d the new variable for the overall score. It can be used to find out the value of the MEAN by (q1, q2,...) formula.

Figure 3: Histogram of total score

(Source: Self-Created in SPSS)

The histogram has been created and the mean of 49 and the standard deviation is 3.712, the histogram shows at the level of 50 total score frequency is very high.

Stage 2 Eliminating poorly performing items: Item analysis & reliability

Test bias

For finding the extremely distorted or poorly Scattered Items it should be examined the item distributions visually and statistically. If the distribution is lopsided or unevenly spread, eliminate a single thing at a time. After each elimination, it must be recalculated the overall scale score. Then in the next stage, any potential demographic biases can be observed (Stanley, et al. 2021). Then Eliminate biased elements to ensure equality between groups.

Inter-item correlation

From the extremely Strong Inter-Item Correlations, the inter-item correlations can be observed to find items having these relationships. The remaining items' reliability is assessed cautiously in order to preserve the ideal equilibrium between item elimination and scale accuracy (Roberts, et al. 2021).

Item-total correlation

In the next step, it should reevaluate associations after eliminating a single thing at a time. Poor Item-Total Correlations are examined for the relationships between specific items and the scale's overall score (Ghazali, et al. 2021). Items are discarded with weak or negative associations one at a time, modifying the scale's overall score in the process.

Reliability

Then the Verification of the Reliability of the Last Items the remaining elements, can perform a reliability analysis (such as Cronbach's Alpha). If the scale's accuracy is sufficient, it might want to stop eliminating any more items. It can be noted the elements that might be removed to boost reliability even more (Pei-Boon, et al. 2020). Inadequate Reliability can be kept by eliminating items one at a time if the validity of the scale is lacking. After each elimination, the reliability should be evaluated.

Big-5 (Cronbachs Alpha)

Figure 4: Neuroticism Cronbachs alpha

(Source: Self-Created in SPSS)

The internal coherence or reliability of a scale is measured by Cronbach's alpha, which shows how effectively the items measure the same concept. 0.7 Or greater is a typical cutoff for satisfactory dependability, however, this might change based on the context and goal of the study. It can be seen that there are 10 items on the scale for neuroticism, and the real Cronbach's alpha is 0.5656- a lower number than the 0.84 that was stated. The items may not assess the same concept of neuroticism because they are not highly consistent with one another (Roberts, et al. 2021). The actual Cronbach's alpha may not match the reported Cronbach's alpha because of differences in participant samples. Sampling error is the discrepancy between the population parameter and sample statistic that results from random fluctuation. The Cronbach's alpha that has been published might not be the same as the alpha that has actually been measured. The discrepancy between the observed value and the real value as a result of biases or mistakes in the measuring process is known as measurement error [Referred to appendix 2].

Stage 3: Evaluating the scales psychometric properties: Discrimination indices & validity

Corrected item-total correlations

At first, the Items are listed which are Most and Least Discriminating and for each item on your scale, it should be calculated the corrected item-total correlations. The listed items with the greatest and least significant relationships can produce scatterplots (Pei-Boon, et al. 2020). For producing the scatterplots for the items with the highest and lowest discriminatory it is used with the adjusted item-total correlations and can be subtracted from the evaluations for each individual item from the scale's overall score [Referred to appendix 3].

Method of extreme groups

Figure 5: The top and lowest 25% of scorers on the major build scale using a variable

(Source: Self-created in SPSS)

For creating a variable that will show the most elevated and bottom twenty-five per cent of your primary build scale's scores. It can determine the means for the top and worst 25% of scores for each scale item. Then the variations of scores in averages for each item should be noted. then the total scale score is updated using only the surviving main construct elements, taking into account the results of the discrimination analysis (Stanley, et al. 2021) Correlations between the primary construct scale score and the reliability of the validity test, measure score, as well as correlations between specific items, are used to evaluate the scale's validity. If pertinent data are available, validity examinations, such as known categories or criterion-related measurements, are carried out.

Validity

Figure 6: Correlation for validity check

(Source: Self-created in SPSS)

It must compare the modified total scale score to the overall scale score for the Big-5 construct validity check. It must conduct further validity analyses like correlations, t-tests depending on the data gathered such as, known groups, criteria validity. Then, a table with the total correlations of items for each item is updated. Incorporate scatterplots are obtained for the items that are most and least discriminating (Purwanto, 2021). It produces a table showing the values for every scale component for the most extreme groups, emphasizing differences. It analyzes the means table to find categories that effectively distinguish between extreme groupings [Referred to appendix 1].

Identification of the most and least discriminating items in the scale

Figure 7: Identification of the most and least discriminating items in the scale through correlation matrix

(Source: Self-Created in SPSS)

Setting high expectations for myself and others, which has the greatest average inter-item correlation of 0.133, is the most discriminating item on the scale. This indicates that this item is most in line with the overall construct that the scale is measuring. Demand perfection in others, which has the lowest average inter-item correlation of -0.125, is the least discriminating item on the scale (Nasution, Harahap, & Ritonga, 2022). This indicates that this item is the least consistent with the broad construct that the scale is measuring.

Scatter plot between setting high expectations for myself and others and Demanding perfection from others

Figure 8: Scatter plot between setting high expectations for myself and others and Demanding perfection in others

(Source: Self-created in SPSS)

The graphic depicts a dispersed diagram of individual expectations for excellent quality and holding oneself and others to exacting standards. Setting high standards is shown on the y-axis while demand for quality is represented on the x-axis. The data sets are colour-coded, with the strong agreement being represented by red, agreement by orange, neutrality by yellow, disagreement by green, and strong disagreement by blue (Stanley, et al. 2021)). A positive association between demanding quality and establishing high standards may be seen in the scatter plot. This implies that those who demand excellent service are more inclined to hold themselves and others to high standards. This is probably because those who seek excellent quality tend to be more critical of both their own and other people's work.

Stage 4: Descriptive statistics & norms

Descriptive statistics

Figure 9: Descriptive stat of main construct scale

(Source: Self-created in SPSS)

The Frequencies module provided percentiles and quartiles, which helped present a comprehensive view of the score distribution. This meticulous procedure guarantees a thorough study of the central patterns, accuracy, and distribution of our major construct scale, providing valuable data (Purwanto, 2021). The table exhibits that everything's answers fluctuated essentially from each other. For example, the reaction "Going with the ideal choice is fundamental for me" gets a standard deviation of 1.1, while the reaction "Don't consider an errand finished unit's ideal" has a standard deviation of 1.4. This shows that individuals' thoughts on what is vital for them could go incredibly and that they can vary altogether also.

Standard Error of Measurement & Confidence Intervals

Figure 10: Reliability for standard Error measurement

(Source: Self-created in SPSS)

A reliability test for Cronbachs alpha is calculated as the (-0.714). The mean is 49, the variance is 13.778 and the standard deviation is 3.712. A table of dependability data for a scale that measures standard error is seen in the image (Roberts, et al. 2021). According to the table, Cronbach's alpha is -0.714. Given that this number is negative, the scale is inaccurate. A low Cronbach's alpha could occur for a number of causes. One option is that the scale's elements measure various things.

Standard error of measurement

SEM = 3.712√ (1-0.714)

SEM = 3.712 X 0.5347896 = 1.985139

Figure 11: Confidence Intervals (CIs) of 68%, 95%, and 99% for a particular score

(Source: Self-created in SPSS)

It first acquired essential measurements like the mean, deviation from the mean, and median via the descriptives module in SPSS for the examination of descriptive data and guidelines for the main construct scale. In the next step, it can be calculated the Standard errors of Measurement (SEM) for the scales, which is essential for comprehending the accuracy of our measurements, in order to improve the understanding of our results (Patil, & Modi, 2019). To provide a wide range around a given score, we also painstakingly computed and provided 68%, 95%, as well as 99% confidence intervals (CIs). Alongside distinct measurements for the essential building scale.

Norms table

Score	Z-Score	T-Score	Percentile	Quartile
10	-1.67	33.3	4.76
12	-1.17	38.3	12.38
14	-0.67	43.3	24.88	25th
16	-0.17	48.3	43.14
18	0.33	53.3	63.29
20	0.83	58.3	79.69
22	1.33	63.3	90.82
24	1.83	68.3	96.62

Table 12: Norms table for Z score, T score, percentile, Quartile

(Source: Self-created in SPSS)

A standards table with z-score t-scores, percentiles, and quartiles is made, giving an exhaustive image of the scale's mental qualities. This type of examination is a complex multistage approach that guarantees the appraisal device has been thoroughly assessed and approved. It created a norms table with t and z-scores, quartiles and percentiles to improve the understanding of the distribution of results (Alita, Putra, & Darwis, 2021). The Transform module was used to compute the z-scores and t-scores, with the latter resulting from the former under the assumption of standard Z-scores.

Conclusion

In conclusion, it can be ensured that the validity and reliability of psychological tests depends critically on the study of a psychologically sound measure. The study gets a thorough grasp of the evaluation instruments quality by carefully studying aspects such as item distributions and biases, inter-item connections and reliability. Measurement accuracy is improved through the repeated process of scale refinement through item reduction, reliability evaluations, and discrimination indices. A strong foundation for inferring meaningful conclusions from the data is provided by the assessment of psychometric qualities, which includes statistical analysis, norms, and validity checks. The accuracy and usefulness of psychological assessments in study and application are eventually advanced by a psychometrically sound measure, which also contributes to the continual improvement of assessment tools.

References

Journals

• Alita, D., Putra, A. D., & Darwis, D. (2021). Analysis of classic assumption test and multiple linear regression coefficient test for employee structural office recommendation. IJCCS (Indonesian Journal of Computing and Cybernetics Systems), 15(3), 295-306.Retrieved From: https://journal.ugm.ac.id/ijccs/article/download/65586/31843 [Retrieved On: 29.09.2023]

Books

• El-Den, S., Schneider, C., Mirzaei, A., & Carter, S. (2020). How to measure a latent construct: Psychometric principles for the development and validation of measurement instruments. International Journal of Pharmacy Practice, 28(4), 326-336.Retrieved From: https://onlinelibrary.wiley.com/doi/pdfdirect/10.1111/ijpp.12600 [Retrieved On: 29.09.2023]
• Ghazali, N., Mustakim, S. S., Nordin, M. S., & Hashim, S. (2021). Assessing the Psychometric Properties of Students' MOOC-Efficacy Measurement Model. Pertanika Journal of Social Sciences & Humanities, 29.Retrieved From: https://www.researchgate.net/profile/Norliza-Ghazali/publication/350898795_Assessing_the_Psychometric_Properties_of_Students'_MOOC-Efficacy_Measurement_Model/links/60d19d6d299bf1cd71e9b5bd/Assessing-the-Psychometric-Properties-of-Students-MOOC-Efficacy-Measurement-Model.pdf [Retrieved On: 29.09.2023]
• Hidalgo, M. A., Petras, H., Chen, D., & Chodzen, G. (2019). The Gender Minority Stress and Resilience Measure: Psychometric Validity of an Adolescent Extension. Clinical Practice in Pediatric Psychology, 7(3), 278.Retrieved From: https://psycnet.apa.org/manuscript/2019-52280-007.pdf [Retrieved On: 29.09.2023]
• Howitt, D., & Cramer, D. (2020). Understanding statistics in psychology with SPSS. Pearson.Retrieved From: https://thuvienso.hoasen.edu.vn/bitstream/handle/123456789/12855/Contents.pdf?sequence=1 [Retrieved On: 29.09.2023]
• Nasution, A. A., Harahap, B., & Ritonga, Z. (2022). The Influence of Product Quality, Promotion and Design on Purchase Decisions for Yamaha Nmax Motor Vehicles SPSS Application Based. International Journal of Economics (IJEC), 1(1), 01-13.Retrieved From: https://ejournal.ipinternasional.com/index.php/ijec/article/download/67/45 [Retrieved On: 29.09.2023]
• Ntumi, S., Agbenyo, S., & Bulala, T. (2023). Estimating the Psychometric Properties (" Item Difficulty, Discrimination and Reliability Indices") of Test Items Using Kuder-Richardson Approach (KR-20). Shanlax International Journal of Education, 11(3), 18-28.Retrieved From: https://files.eric.ed.gov/fulltext/EJ1388606.pdf [Retrieved On: 29.09.2023]
• Patil, V., & Modi, M. (2019). 3rd Party logistics providers: comparative study and analysis by using SPSS in Indian market. Int J Sci Technol Res, 8(9), 2010-2018.Retrieved From: https://d1wqtxts1xzle7.cloudfront.net/63417613/3rd-Party-Logistics-Providers-Comparative-Study-And-Analysis-By-Using-Spss-In-Indian-Market20200525-84494-1ur72yq-libre.pdf?1590400663=&response-content-disposition=inline%3B+file%3D3_rd_Party_Logistics_Providers_Comparati.pdf&Expires=1696679133&Signature=Vbl5Ml-XHKotiK90OcHiEHSbwUyKyVvjN7-xEmubWzucGkwZNCuHxfLC85HLO4JacHHXgomagn6FTAhDwGnM9HfjMBj48cZBEHR57nVDNUqXw~20340SH6W~ajUXTgX4~1gq2W4DVDP6YYCgwo6M86ou8qAkLDYldcX2JMRDGJM46y2cSrUveEaPiZcBXUDaqqDFMMmrS61PlCyveZe-NKCIfFPb3ATBTH6lipysOhxz2CZxvzMYyEP~ZhlqJxss~mytHQJuAuLWFpvINVsbozMhBymI26iP6rBkMLOnVjFXcsRC3w-rv57lnEbz6SesayxQkqBQTU~vCPDq2Gw-hQ__&Key-Pair-Id=APKAJLOHF5GGSLRBV4ZA [Retrieved On: 29.09.2023]
• Pei-Boon, O., Wan Jaafar, W. M., Chin-Siang, A., & Nee-Nee, C. (2020). Psychometric properties of the sources of counseling self efficacy in a sample of Malaysian secondary school counselors. SAGE Open, 10(1), 2158244020902076.Retrieved From: https://journals.sagepub.com/doi/pdf/10.1177/2158244020902076 [Retrieved On: 29.09.2023]
• Purwanto, A. (2021). Education Management Research Data Analysis: Comparison of Results between Lisrel, Tetrad, GSCA, Amos, SmartPLS, WarpPLS, and SPSS For Small Samples. Nidhomul Haq: Jurnal Manajemen Pendidikan Islam, 6(2).Retrieved From: https://www.e-journal.ikhac.ac.id/index.php/nidhomulhaq/article/download/1575/683 [Retrieved On: 29.09.2023]
• Roberts, N. P., Kitchiner, N. J., Lewis, C. E., Downes, A. J., & Bisson, J. I. (2021). Psychometric properties of the PTSD Checklist for DSM-5 in a sample of trauma exposed mental health service users. European Journal of Psychotraumatology, 12(1), 1863578.Retrieved From: https://www.tandfonline.com/doi/pdf/10.1080/20008198.2020.1863578 [Retrieved On: 29.09.2023]
• Stanley, I. H., Hom, M. A., Christensen, K., Keane, T. M., Marx, B. P., & Björgvinsson, T. (2021). Psychometric properties of the Depressive Symptom Index-Suicidality Subscale (DSI-SS) in an adult psychiatric sample. Psychological Assessment, 33(10), 987.Retrieved From: https://psycnet.apa.org/manuscript/2021-53180-001.pdf [Retrieved On: 29.09.2023]
• Tournaki, N., Woodcock, S., & Ehrich, J. (2023). Tale of the tape: psychometric investigation of two popular teacher self-efficacy scales. Teaching Education, 1-18.Retrieved From: tandfonline.com/doi/pdf/10.1080/10476210.2023.2239714 [Retrieved On: 29.09.2023]