4.1. Common method deviation test
Common method deviation refers to the artificial covariation among variables caused by the same data source, measurement environment, project context, and project characteristics. The common method deviation will lead to inaccurate data and misleading conclusions, so it should be controlled. Generally, the common method deviation is controlled by both procedures and statistics.
Procedural control: (l) When the questionnaire is applied, the subjects are asked to answer anonymously, and the guidelines of the questionnaire and the subjects' answers emphasize the anonymity and confidentiality of the research; (2) By choosing subjects to answer in different schools, places and periods, the measurement is separated in time and space; (3) Set a certain number of reverse scores to prevent the subjects from answering carelessly, and try to detect the true level of the subjects as much as possible.
Although the program is controlled to a certain extent, it is still difficult to eliminate the common method deviation. Therefore, this study further uses Harman single factor test to make exploratory factor analysis on all measurement items and finds that there are 26 items with factor characteristic values greater than 1, and the maximum factor explanation variance is 15.540%, less than 40%. Therefore, it can be considered that the common method deviation is well controlled in this study, and there is no common method deviation problem.
4.2. Descriptive statistic
In this study, perfectionism, career adaptability, and career decision-making difficulties are analyzed utilizing mean, standard deviation, maximum value, and minimum value, among which perfectionism is divided into positive perfectionism and negative perfectionism. As shown in the following Table 1.
Table 1
Descriptive statistics of positive perfectionism, negative perfectionism, career adaptability, and career decision-making difficulties.
Items | Min | Max | M | SD |
Positive perfectionism | 1.21 | 5.00 | 3.75 | 0.74 |
Negative perfectionism | 1.53 | 5.00 | 2.83 | 0.58 |
fear of mistakes | 1.00 | 5.00 | 2.31 | 0.61 |
parents' expectation | 1.00 | 5.00 | 2.74 | 0.81 |
hesitation in actions | 1.00 | 5.00 | 3.12 | 0.75 |
personal standards | 1.00 | 5.00 | 3.21 | 0.78 |
Career Adaptability | 1.62 | 4.95 | 3.81 | 0.52 |
career control | 1.80 | 5.00 | 3.81 | 0.49 |
career curiosity | 1.23 | 5.00 | 3.73 | 0.64 |
career concern | 1.64 | 5.00 | 3.48 | 0.65 |
career confidence | 1.33 | 5.00 | 3.79 | 0.57 |
career adjustment | 1.31 | 5.00 | 3.71 | 0.58 |
career interpersonal relationship | 1.20 | 5.00 | 3.12 | 0.65 |
Career Decision-Making Difficulties | 1.33 | 4.28 | 2.71 | 0.57 |
lack of preparation | 1.00 | 4.87 | 2.59 | 0.64 |
difficulty in information exploration | 1.00 | 4.96 | 2.72 | 0.71 |
conflict and contradiction | 1.00 | 4.77 | 2.76 | 0.57 |
In terms of perfectionism. The average value of positive perfectionism is significantly higher than the median value, which indicates that the level of positive perfectionism of college students is at a high level. The total score of negative perfectionism and the scores of fear of mistakes and parents' expectations are lower than the median, but the scores of personal standards and hesitation in actions are higher than the median. After entering the university, college students have more initiative and autonomy in their study and life and pursue organization and order in doing things. At the same time, due to further separation from parental control and the influence of parental expectations on college students gradually decreases, which led students to have high self-confidence and will not deny themselves because of a temporary failure. However, college students can't set their own personal standards according to the actual situation. Moreover, due to their lack of ability and the inability to see a longer-term future, they can't predict the mistakes and failures they will face, and they will have doubts when they act.
In terms of career adaptability. The scores of career adaptability in general and all dimensions are higher than the median, which indicates that college students' career adaptability is at a high level. Nowadays, universities no longer only emphasize grades, which makes college students have more extracurricular activities after studying, so as to get on well with classmates and cultivate their own interests and hobbies. In addition, the popularity of the Internet also enables college students to understand the latest information in all aspects faster, which is conducive to enhancing their self-confidence and promoting their good interpersonal relationships, and ultimately improving their career adaptability. However, the score of the career control dimension is low. The reason is that most college students learn books mechanically, and have little or no chance to do social practice related to their own majors. When they need to do something, they will feel at a loss, anxious, and uncontrollable.
In terms of career decision-making difficulties. This study found that the average values of career decision-making difficulties in general and all dimensions are slightly lower than the median, which indicates that contemporary college students have some degree of career decision-making difficulties. Specific performance can be classified into three points: lack of awareness and relevant knowledge of career planning, inability to prepare for career planning in advance, and always think that career choice should be considered by college students who are nearing graduation.
4.3. Correlation analysis
Pearson correlation coefficient can well test the correlation between two variables [53]. In this study, the Pearson correlation coefficient was used to test the relationship among positive perfectionism, negative perfectionism, career adaptability, and career decision-making difficulties. The specific results are shown in Table 2 below.
Table 2
| 1 | 2 | 3 | 4 |
1. Positive perfectionism | - | | | |
2. Negative perfectionism | 0.185** | - | | |
3. Career Adaptability | 0.403** | -0.084* | - | |
4. Career Decision-Making Difficulties | -0.139** | 0.315** | -0.502** | - |
Note: *: p < 0.05; **: p < 0.01 |
The relationship between the four variables shows a significant pairwise correlation. Positive perfectionism is positively correlated with career adaptability (r = 0.182, 0.406; P < 0.01), which was negatively correlated with career decision-making difficulties (r=-0.140; P < 0.01). Negative perfectionism is negatively correlated with career adaptability (r=-0.083; P > 0.05), which was positively correlated with career decision-making difficulty (r = 0.317; P < 0.01). There is a significant negative correlation between career adaptability and career decision-making difficulty (r=-0.487; P < 0.01)。
4.4. Regression analysis of perfectionism and career decision-making difficulties
Based on the correlation analysis of positive perfectionism, negative perfectionism, career adaptability, and career decision-making difficulties, this paper constructs a regression model of positive perfectionism, negative perfectionism, and career decision-making difficulties combined with theoretical analysis, and investigates the predictive effects of positive perfectionism and negative perfectionism on career decision-making difficulties of college students. The results are shown that X2/df = 3.186, RMSEA = 0.075, NFI = 0.952, IFI = 0.964, CFI = 0.951, GFI = 0.973, AGFI = 0.969, which indicates that this model has a good fitting degree.
Besides, the main path coefficients of the regression model show that the standardized regression coefficient of positive perfectionism to career decision-making difficulties is -0.171, P < 0.000. The standardized regression coefficient of negative perfectionism to career decision-making difficulties is 0.479, P < 0.000. The regression path coefficient differences are extremely significant, which shows that positive perfectionism can negatively predict career decision-making difficulties, and college students with higher positive perfectionism can reduce the occurrence of career decision-making difficulties. Negative perfectionism can positively predict career decision-making difficulties, while higher negative perfectionism of college students is not conducive to the smooth production of career decision-making.
4.5. Mediating effect analysis
According to Hayes and Scharkow point out that mediation analysis can effectively analyze the influence of variable X on variable Y, which makes the research progress in statistical methods and obtain more in-depth research results [54]. In addition, if the independent variable x has a certain influence on the dependent variable y through the variable M, then the variable m is the intermediary variable of X and Y, and the specific mediating effect can be described by the regression equation path diagram in Fig. 3.
Based on the above model, this study uses Amos 22.0 software to test the mediating effect of career adaptability on perfectionism and career decision-making difficulties. The results are shown that X2/df = 2.564, RMSEA = 0.069, NFI = 0.948, IFI = 0.959, CFI = 0.963, GFI = 0.968, AGFI = 0.971, which indicates that this model has a good fitting degree.
Furthermore, this study analyzes the main path coefficients of the mediation model of perfectionism, career adaptability, and career decision-making difficulties of college students, and the results are shown in the following Table 3.
Table 3
Main path coefficients of the mediation model.
Path | Estimate | SE | CR | P |
NP◊CA | -0.711 | 0.054 | -3.174 | 0.001 |
PP◊CA | 0.298 | 0.048 | 6.664 | 0.000 |
CA◊CDMD | -0.589 | 0.087 | -6.989 | 0.000 |
NP◊CDMD | 0.331 | 0.059 | 5.411 | 0.000 |
PP◊CDMD | -0.026 | 0.046 | -0.608 | 0.243 |
From the above data, it can be seen that the fitting index of this model, except for the influence of positive perfectionism on career decision-making difficulties, does not meet the standard, and other fitting indexes all meet the standard, so the model fits well. It shows that career adaptability plays an intermediary role in the role of positive perfectionism and negative perfectionism in career decision-making difficulties. Specifically, career adaptability plays a partial mediating role in the relationship between negative perfectionism and career decision-making difficulties. It plays a complete intermediary role in the relationship between positive perfectionism and career decision-making difficulties, and the research hypothesis holds. Therefore, this study establishes a mediation model, as shown in the following Fig. 4.
Negative perfectionism directly affects the career decision-making level of college students, the higher the score of negative perfectionism of college students, the easier it is to make career decision-making difficulties. Negative perfectionism can also influence career decision-making difficulties through career adaptability, if college students have higher negative perfectionism, it will be detrimental to the development of career adaptability, and college students with lower career adaptability are more likely to have career decision-making difficulties.
Positive perfectionism does not directly affect career decision-making difficulties but exerts influence on career decision-making difficulties completely through career adaptability. College students' higher positive perfectionism is conducive to the development of their career adaptability, and the improvement of career adaptability helps college students make career decisions smoothly, resulting in fewer career decision-making difficulties.
After fitting the mediation analysis model, it is necessary to decompose the mediation effect. Usually, the contents of decomposition mainly include: (1) the size of mediation effect; (2) The ratio of intermediary effect to total effect, namely ab/ (ab + c'); (3) The ratio of intermediary effect to direct effect, namely ab/c'; (4) Analyze the size of the specific mediating effect, that is, the total mediating effect through a certain mediating variable.
As shown in Table 4, in the mediation model of negative perfectionism, career adaptability, and career decision-making difficulties, the direct effect is 0.334, the mediating effect is 0.099, and the total effect is 0.433. In the mediation model, ab and c' have the same sign and are significant, which indicates that the mediation is part of the mediation, and the mediation effect accounts for 22.9% of the total effect.
Table 4
Effect analysis of negative perfectionism.
Effect | Effect value | Result |
Direct Effect | c’ | 0.334 |
Mediating Effect | ab | 0.099 |
Total Effect | ab + c’ | 0.433 |
The ratio of mediating effect to total effect | ab/(ab + c’) | 0.229 |
The ratio of mediating effect to direct effect | ab/ c’ | 0.296 |
As shown in Table 5, in the mediation model of positive perfectionism, career adaptability, and career decision-making difficulties, the direct effect is -0.032, the mediation effect is -0.244, and the total effect is -0.276. In the mediation model, ab and c' have the same sign, but c' is not significant, which indicates that the mediation model is complete mediation, and the mediation effect accounts for 88.4% of the total effect.
Table 5
Effect analysis of positive perfectionism.
Effect | Effect value | Result |
Direct Effect | c’ | -0.032 |
Mediating Effect | ab | -0.244 |
Total Effect | ab + c’ | -0.276 |
The ratio of mediating effect to total effect | ab/(ab + c’) | 0.884 |
The ratio of mediating effect to direct effect | ab/ c’ | 7.625 |